Market Returns

Highlights The COVID-19-induced demand shock in China – and a stronger USD – will reduce growth in global crude oil consumption to just over 1mm b/d this year, vs. earlier expectations of ~ 1.4mm b/d. Significant fiscal and monetary stimulus from China will be required to put economic growth back on track over the critical 2020-21 interval. An accommodative monetary-policy backdrop globally also will support demand. On the supply side, OPEC 2.0 likely will cut output by an additional 600k b/d in 2Q20, which will remove 2.3mm b/d off member states’ official quotas. For 2H20, we expect the coalition to revert to its 1.7mm b/d in cuts to keep markets balanced. US shale-oil output growth will continue to slow under market-imposed capital discipline. We are revising our baseline price forecasts in 2020 lower to $62/bbl and $58/bbl for Brent and WTI, respectively (Chart of the Week). This is down $5/bbl vs our previous forecast. Price risk is to the upside, however. 2021 Brent and WTI forecasts remain at $70/bbl and $66/bbl, respectively, as we do not expect long-lived demand destruction from the COVID-19 outbreak. A growing consensus around policy stimulus and production cuts makes us leery. Feature Chart of the WeekCOVID-19 Knocks Oil Forecasts Lower COVID-19 continues to hammer Chinese oil demand, forcing refiners there to drastically reduce output. This crude oil is ending up in inventories, but, so far at least, overall storage capacity in China is not being maxed out by the unintended accumulations of crude and product inventories. Data are difficult to come by, but there are a few observations that provide some insight into the state of the refining market in China as the COVID-19 episode unfolds. Platt’s reported independent refiners in Shandong Province, which has ~ 3.4mm b/d of refining capacity, cut runs to a four-year low of ~ 40% of capacity this month, down from a January rate of 63.5%. Shandong refiners represent 50%-60% of China’s independent refining capacity.1 We estimate EM demand – led by downward revisions in China – will fall by ~900k b/d in 1Q20 – when most of the damage to the economy likely will occur – and by an average 300k b/d for the year vs. our previous estimates. Ursa Space Systems’ radar satellite monitoring of inventories close to coastal refineries indicated Chinese oil storage at the beginning of the month was at 60% of capacity.2 This figure likely is higher, given refinery runs remain low, but it does not yet suggest storage capacity in China will be exhausted in the near future. In our modeling of the COVID-19 impact on oil demand, we estimate EM demand – led by downward revisions in China – will fall by ~900k b/d in 1Q20 – when most of the damage to the economy likely will occur – and by an average 300k b/d for the year vs. our previous estimates. This leads us to believe EM oil demand will increase by 1mm b/d this year, down from our earlier expectation of 1.26mm b/d pre-COVID-19. For DM economies, demand growth also will disappoint, revised down by 100k b/d on the back of a warmer-than-expected winter and stop-and-go growth in manufacturing induced by COVID-19. Policy Stimulus Will Revive Chinese Demand The COVID-19 outbreak will result in a significant hit to China’s GDP, which will require substantial stimulus to put growth back on a 6% p.a. track this year. This growth rate is required for the Chinese Communist Party (CCP) to deliver on its pledge to double GDP and per-capita income over 2010-20, a pledge that was memorialized in writing following the Party’s 2012 Congress. In addition, next year marks the 100th anniversary of the founding of the CCP, and, we believe, it is an all-but-foregone conclusion the Party’s leadership will not want a faltering economy on display as it celebrates this important milestone. Given these considerations, the possibility policymakers will over-stimulate the economy to get it back on track is a non-trivial upside risk.3 We do not think it is unreasonable to expect policymakers to lean into reviving growth this year and next with policy stimulus. Our baseline 2020 forecast envisions prices will falter somewhat versus our previous expectation – with Brent averaging $62/bbl this year, and WTI trading $4/bbl below that, vs. $67/bbl and $63/bbl previously. We are mindful of the impact Chinese policy stimulus can have on the global oil markets. The effects on GDP growth following demand shocks of past stimulus can be seen in the response of China’s GDP following the 2003 SARS outbreak; the 2008-09 GFC; the 2011-12 eurozone debt crisis; and even in China’s 2015-16 slowdown (Chart 2). For this reason, we do not think it is unreasonable to expect policymakers to lean into reviving growth this year and next with policy stimulus. And it is for this reason that we believe price risk tilts to the upside this year. Our updated Ensemble price forecast includes two additional demand-side simulations to assess its sensitivity to changes in EM oil demand: Chart 2Chinese Stimulus Will Support Oil Demand Higher EM demand scenario (20% weight): We model the impact of the coronavirus as short-lived, with only a temporary impact on China’s economy. Consumer demand and industrial production in China converge to pre-COVID-19 levels rapidly in 2H20. Chinese policymakers overstimulate in 2Q20, over fears the virus could have severe long-term consequences on the economy. This scenario assumes EM demand increases by 100k b/d vs. our base case in 2020 and 2021. Lower EM demand scenario (10% weight): We model the impact of the coronavirus as a severe and long-lasting event. This triggers a negative feedback loop for EM oil demand; collapsing demand forces production lower, which reduces employment and pushes demand further down. This reverberates to other EM economies and affects global supply chains. This scenario assumes EM demand decreases by 240k b/d in 2020 and returns to our base case in 2021, supported by China stimulus. Oil-Demand Reduction (Not Destruction) The outbreak also is contributing to greater global economic uncertainty, which continues to support the USD broad trade-weighted index (TWIB). The COVID-19 outbreak in China caused us to reduce our expectation for global oil demand growth by ~ 360k b/d, taking 2020 year-on-year growth to ~ 1.04mm b/d, versus our earlier expectation of 1.4mm b/d. The outbreak also is contributing to greater global economic uncertainty, which continues to support the USD broad trade-weighted index (TWIB). Dollar strength produces a headwind for EM GDP growth, which suppresses oil-demand growth. The combination of the COVID-19-induced demand reduction and the stronger USD TWIB likely will compel OPEC 2.0 to maintain its production discipline until the global policy uncertainty abates and the USD TWIB retreats. Such a reversal in trend would become a tailwind for commodity demand (Chart 3). Chart 3Global Economic Uncertainty Keeps A Bid Under USD TWIB Global supply growth will continue to be constrained by demands from investors to return capital to shareholders. We expect the hit to global demand to be offset by increased production cuts from OPEC 2.0, which will be agreed next month. OPEC 2.0 production also will be impacted by continued output losses in Iran and Venezuela, which have seen y/y production fall by ~ 1.8mm b/d in 2019. Global supply growth will continue to be constrained by demands from investors to return capital to shareholders – via stock buybacks – and for steady and increasing dividends to make their equity competitive with alternative sectors (e.g., tech). These capital-market pressures – in addition to growing pressure from Environmental, Social and Governance (ESG) investors – will continue to have a profound effect on capital availability for oil and gas E+P companies for decades to come. This is a theme we will return to often in future research. We summarize these supply-demand dynamics in Chart 4. For OPEC 2.0, the 1.7mm b/d reduction in output the coalition agreed for 1Q20 remains in place, as do losses from Iran and Venezuela. For 2Q20, we assume the coalition adds another 600k b/d of production cuts. After that, we assume OPEC 2.0 reverts to its earlier production cuts of 1.7mm b/d for 2H20. In 2021, we assume OPEC 2.0 takes production cuts back down to 1.2mm b/d in January 2021, then gradually increases its production over 1H21 to balance the market and to avoid spiking prices. We also expect the Kingdom of Saudi Arabia (KSA) to remove 300k b/d of overcompliance next year, as markets tighten. In 2H21, we see OPEC 2.0 production levels remaining flat at ~ 44.8mm b/d (Table 1). Table 1BCA Global Oil Supply - Demand Balances (MMb/d, Base Case Balances) Chart 4Supply-Demand Balances Chart 5Global Oil Inventories Will Resume Drawing For 2021, we are leaving our Brent forecast unchanged at $70/bbl, and WTI at $66/bbl. For the US, we reduced our Lower 48 production assumptions, and now have 740k b/d growth in 2020 and 300k b/d in 2021. Shales account for almost all of this increase. We also include a scenario in which US production comes in lower in our ensemble forecast. These fundamentals combine to put global oil inventories back on a downward trajectory in 2H20 (Chart 5). That said, there is an important caveat going into 2H20: If the US Economic Policy Uncertainty Index starts rising in 2H20 on the back of US election risks, markets will continue to price in a stronger USD in 2020 vs. what we now expect. For 2021, we are leaving our Brent forecast unchanged at $70/bbl, and WTI at $66/bbl. Odds favor a return to the pre-COVID-19 price trajectory for oil next year, with continued upside risk from Chinese fiscal and monetary stimulus, and a globally accommodative monetary-policy backdrop. Higher Spare Capacity Reduces Risk Premium The market remains partly balanced by OPEC 2.0’s production cuts. This means that the group’s spare capacity is increasing, reducing the risk premium the market typically includes in crude oil prices to reflect sudden output losses. The risk premium in oil prices evaporated following the drop in demand and the increase in spare capacity due to the large OPEC 2.0 cuts. When China’s economy resumes its normal activity, demand will pick up and the market will balance, increasing the impact of possible supply disruptions. However, the market remains partly balanced by OPEC 2.0’s production cuts. This means that the group’s spare capacity is increasing, reducing the risk premium the market typically includes in crude oil prices to reflect sudden output losses. In addition, if production capacity of ~ 300k-500k b/d in the Neutral Zone shared by KSA and Kuwait is restored, the risk premium could drop even lower, given this production is expected to be retained as spare capacity. If this is the case we could have lower prices in 2020 vs. our current forecast (down to ~ $60/bbl). We will be exploring the changes in OPEC 2.0 spare capacity and the consequences for overall production in future research. Bottom Line: Assisted by Chinese policy stimulus, oil demand will recover this year from the COVID-19-induced demand shock. On the supply side, the combination of deeper OPEC 2.0 production cuts – which we expect will be settled at the upcoming March meeting – and capital-market-imposed reduction in US oil production will push oil markets to a supply deficit. The ongoing demand shock forces us to reduce our 2020 Brent price forecast to $62/bbl from $67/bbl previously. For 2021, we maintain our $70/bbl target. Risks to our view are mounting. Three crucial pieces to our 2020 and 2021 expectations remain uncertain: The duration and magnitude of the impact of the coronavirus shock, The level of production cuts by OPEC 2.0 and the degree of compliance by all members, and The trajectory of the US dollar – if global economic policy uncertainty remains elevated the USD could remain well bid, which would continue to pressure EM GDP growth – and commodity demand – at the margin. Our base case remains that prices will rise from here, but our conviction level is slightly lower. One reason for this is the apparent consensus emerging around the likelihood of Chinese stimulus and OPEC 2.0 production cuts. If either of these assumptions prove wrong, oil prices likely would move lower.   Robert P. Ryan Chief Commodity & Energy Strategist rryan@bcaresearch.com Hugo Bélanger Associate Editor Commodity & Energy Strategy HugoB@bcaresearch.com     Commodities Round-Up Energy: Overweight As of Tuesday’s close, Brent prices were up 8% from their Feb 10 low of $53.27/bbl, supported by receding COVID-19 fears and rising expectations OPEC 2.0 will deepen its production cuts at its March meeting. Earlier this week, oil prices received an additional lift from the newly-imposed US sanctions on Rosneft Trading SA – a subsidiary of Russia’s state-own company – for its activities with Venezuela’s PDVSA. Rosneft Trading intensified its involvement in Venezuela’s oil sector and now handles the majority of the country’s crude exports, providing vital support to the Maduro government. The US restrictions include a 90-day wind-down period for companies to end their activities with Rosneft Trading. Base Metals: Neutral Chinese steel consumption – which accounts for ~50% of global demand – has been hit hard by the coronavirus outbreak. Steel and iron ore prices in China plunged 11% and 3% YTD (Chart 6). Steel mills’ inventories increased to record levels, reaching full capacity. Mills are now forced to export their surplus at reduced prices – flooding seaborne steel markets – or to cut output. Accordingly, more than 33% of steel mills are considering cutting steel production, according to a recent Platts survey. Margins at producing mills are declining and could harm high-grade iron ore prices. This is a short-term risk to our view. Precious Metals: Neutral Gold prices surged past $1,600/oz on Tuesday – overlooking positive manufacturing data in the US. Silver shadowed gold’s movement, closing at $18.13/oz. Precious metals are bought as insurance against risks of a wider-than-expected spread of the coronavirus and should remain well bid until uncertainty dissipates. Gold is somewhat overbought based on sentiment, momentum and technical indicators (Chart 7). If, as we expect, the daily increase in confirmed cases ex-Hubei slows meaningfully over the coming months, gold and silver prices will lose some steam. Ags/Softs:  Underweight CBOT March wheat futures surged 4.4% on Tuesday after Australia’s government sharply lowered its estimate of the country’s wheat harvest as severe drought affected crops. The Australian agricultural agency said the crop totaled 15.17 mm MT, the lowest since 2008, paving the way for stronger US exports. Corn also moved higher, with the prompt contract gaining 1.26% on the back of a new round of Chinese tariff exemptions on US goods. A USDA report showed US soybean export inspections bound for China were still half of last year's volumes. Soybeans futures closed 1.25 cents lower at $8.915/bu as markets await large Chinese purchases of US soybeans. Chart 6Increasing Inventories Pressure Steel and Iron ore Prices Chart 7Gold Technical Indicators Signal Overbought Market     footnotes 1     Please see China's Shandong independent refiners cut run rates to 4-year low of 40% in Feb, published by S&P Global Platts February 13, 2020. 2     Please see Oil demand falls on coronavirus: how much will inventories rise? posted by Ursa Space Systems February 7, 2020. 3     Please see Iron Ore, Steel Poised For Rally, published January 13, 2020, for a discussion of the significance of 2020 vis-à-vis the Communist Party’s pledge to double GDP and per-capita income vs. 2010 levels, memorialized by the CCP at its 2012 Peoples Congress.  We also discuss the 100th anniversary of the Party’s founding next year, which also will be a significant milestone for the CCP – and another reason the Party will not want the Chinese economy faltering as it is celebrated.   Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades TRADE RECOMMENDATION PERFORMANCE IN 2019 Q4 Commodity Prices and Plays Reference Table Trades Closed in 2020 Summary of Closed Trades

Highlights Analyses on Asian semis, Argentina and Russia are available on pages 7, 12 and 14, respectively. The most likely trajectory for Chinese growth will be as follows: the initial plunge in business activity will be succeeded by a rather sharp snap-back due to pent-up demand. However, that quick rebound will probably be followed by weaker growth. Financial markets will soon focus on growth beyond the temporary rebound. In our opinion, it will be weaker than markets are currently pricing. Thus, risks for EM risk assets and currencies are skewed to the downside. A major and lasting selloff in EM stocks will only occur if EM corporate bond yields rise. In this week’s report we discuss what it will take for EM corporate credit spreads to widen. Feature The downside risks to EM risk assets and currencies are growing. We continue to recommend underweighting EM equities, credit and currencies versus their DM counterparts. Today we are initiating a short position in EM stocks in absolute terms. Chart I-1 illustrates that the total return index (including carry) of EM ex-China currencies versus the US dollar has failed to break above its 2019 highs, and has rolled over decisively.  In contrast, the trade-weighted US dollar has exhibited a bullish technical configuration by rebounding from its 200-day moving average (Chart I-2). Odds are the dollar will make new highs. An upleg in the greenback will foreshadow a relapse in EM financial markets. Chart I-1EM Ex-China Currencies Have Been Struggling Despite Low US Rates Chart I-2The US Dollar Remains In A Bull Market   Growth Trajectory After The Dust Settles The evolution of the coronavirus remains highly uncertain and unpredictable. As with any pandemic or virus outbreak, its evolution will be complex with non-trivial odds of a second wave. Even under the assumption that the epidemic will be fully contained by the end of March, its economic impact on the Chinese and Asian economies will likely be greater than global financial markets are currently pricing. As investors come to the realization that this initial pick-up in economic activity after the virus outbreak will be followed by weaker growth, the odds of a selloff in equities and credit markets will rise. In our January 30 report titled Coronavirus Versus SARS: Mind The Economic Differences, we argued that using the framework from the SARS outbreak to analyze the current epidemic is inappropriate. First, only a small portion of the Chinese economy was shut down in 2003, and for a brief period of time. The current closures and limited operations are much more widespread and likely more prolonged. Table I-1China’s Importance Now And In 2003 Second, China accounts for a substantially larger share of the global economy today than it did in 2003 (Table I-1). Hence, the global business cycle is presently much more sensitive to demand and production in the mainland than it was during the SARS outbreak. Global financial markets have rebounded following the initial selloff in late January on expectations that the Chinese and global economies will experience a V-shaped recovery. In last week’s report, we discussed why the odds favor a tepid recovery for the Chinese business cycle and global trade. The main point of last week’s report was as follows: with the median company and household in China being overleveraged, any reduction in cash flow or income will undermine their ability to service their debt and will dent their confidence for some time. Hence, consumption, investment and hiring over the next several months will be negatively affected, even after the outbreak is contained. This in turn will diminish the multiplier effect of policy stimulus in China. Chart I-3Our Expectations Of China’s Business Cycle The most likely pattern for Chinese growth will likely resemble the trajectory demonstrated in Chart I-3. It assumes the plunge in business activity will be succeeded by a rather sharp snap-back due to pent-up demand. However, that snap-back will likely be followed by weaker growth, for reasons discussed in last week’s report. Equity and credit markets in Asia and worldwide have been sanguine because they have so far focused exclusively on expectations of a sharp rebound. As investors come to the realization that this initial pick-up in economic activity will be followed by weaker growth, the odds of a selloff in equities and credit markets will rise. Bottom Line: The most likely trajectory for Chinese and Asian growth will be as follows: the initial plunge in business activity will be succeeded by a rather sharp snap-back due to pent-up demand. However, that quick rebound will probably be followed by weaker growth. Financial markets are not pricing in this scenario. Thus, risks are skewed to the downside for EM risk assets and currencies. The Missing Ingredient For An Equity Selloff The missing ingredient for a selloff in EM equities is rising EM corporate bond yields. Chart I-4 illustrates that bear markets in EM stocks typically occur when EM US dollar corporate bond yields are rising. Hence, what matters for the direction of EM share prices is not risk-free rates/yields but EM corporate borrowing costs. Chart I-4The Destiny Of EM Equities Is DependEnt On EM Corporate Bond Yields EM (and US) corporate bond yields can rise under the following circumstances: (1) when US Treasury yields are ascending more than corporate credit spreads are tightening; (2) when credit spreads are widening more than Treasury yields are falling; or (3) when both government bond yields and corporate credit spreads are increasing simultaneously. Provided the backdrop of weaker growth is bullish for government bonds, presently corporate bond yields can only rise if credit spreads widen by more than the drop in Treasury yields. In short, the destiny of EM equities currently relies on corporate spreads. A major and lasting selloff in EM stocks will only occur if their respective corporate bond yields rise. From a historical perspective, EM and US corporate credit spreads are currently extremely tight (Chart I-5). A China-related growth scare could trigger a widening in EM corporate credit spreads. As this occurs, corporate bond yields will climb, causing share prices to plummet. EM corporate spreads have historically been correlated with EM exchange rates, the global/Chinese business cycle, and commodities prices (Chart I-6). The Chinese property market plays an especially pivotal role for the outlook of EM corporate spreads. Chart I-5EM And US Corporate Spread Remain Tame Chart I-6EM Corporate Spreads Inversely Correlate With EM Currencies And Commodities Prices   First, offshore bonds issued by mainland property developers account for a large share of the EM corporate bond index. Chart I-7China Property Market Will Continue Disappointing Second, swings in China’s property markets often drive the mainland’s business cycle and its demand for resources, chemicals and industrial machinery. In turn, Chinese imports of commodities affect both economic growth and exchange rates of EM ex-China. Finally, the latter two determine the direction of EM ex-China corporate spreads. China’s construction activity and property developers were struggling before the coronavirus outbreak (Chart I-7). Given their high debt burden, the ongoing plunge in new property sales and their cash flow will not only weigh on their debt sustainability but also force them to curtail construction activity. The latter will continue suppressing commodities prices. The sensitivity of EM corporate spreads to these variables have in recent years diminished because of the unrelenting search for yield by global investors. As QE policies by DM central banks have removed some $9 trillion of high-quality securities from circulation, the volume of securities available in the markets has shrunk. This has distorted historical correlations of EM corporate spreads with their fundamental drivers – namely, China’s construction activity, commodities prices, EM exchange rates and the global trade cycle. Nonetheless, EM corporate credit spreads’ sensitivity to these variables has diminished, but has not vanished outright. If EM currencies depreciate meaningfully, commodities prices plunge and China’s growth and the global trade cycle disappoint, odds are that EM corporate spreads will widen. Given that credit markets are already in overbought territory, any selloff could trigger a cascading effect, resulting in meaningful credit-spread widening. Bottom Line: A major and lasting selloff in EM stocks will only occur if their respective corporate bond yields rise. The timing is uncertain, but the odds of EM corporate credit spreads widening are mounting as Chinese growth underwhelms, commodities prices drop and EM currencies depreciate. If these trends persist, they will push EM shares prices over the cliff. As to today’s recommendation to short the EM stock index, we anticipate at least a 10% selloff in EM stocks in US-dollar terms. For currency investors, we are maintaining our shorts in a basket of EM currencies versus the dollar. This basket includes the BRL, CLP, COP, ZAR, KRW, IDR and PHP. Arthur Budaghyan Chief Emerging Markets Strategist arthurb@bcaresearch.com   Are Semiconductor Stocks Facing An Air Pocket? Global semiconductor share prices have continued to hit new highs, even though there has not been any recovery (positive growth) in global semiconductor sales or in their corporate earnings (EPS). The coronavirus outbreak and the resulting delay in 5G phone sales in China in the first half of 2020 will trigger a pullback in semiconductor equities. Global semiconductor sales bottomed on a rate-of-change basis in June, but their annual growth rate was still negative in December. In the meantime, global semi share prices have been rallying since January 2019. This divergence between stock prices and revenue of global semiconductor stocks is unprecedented (Chart II-1). Chart II-1Over-Hyped Global Semi Share Prices Odds are that global semi stocks in general, and Asian ones in particular, will experience a pullback in the coming weeks. The coronavirus outbreak will likely dampen expectations related to the speed of 5G adoption and penetration in China. Critically, China accounted for 35% of global semiconductor sales in 2019, versus 19% for the US and 10% for the whole of Europe. In brief, semiconductor demand from China is now greater than the US and European demand combined. Furthermore, the latest news that the US administration is considering changing its regulations to prevent shipments of semiconductor chips to China’s Huawei Technologies from global companies - including Taiwan's TSMC - could hurt chip stocks further. Since Huawei Technologies is the global leader in 5G networks and smartphones, the ban, if implemented, will instigate a sizable setback to 5G adoption in China and elsewhere. Table II-1Industry Forecasts Of The 2020 Global 5G- Smartphone Shipments Our updated estimate of global 5G smartphone shipments is between 160 million and 180 million units in 2020, which is below the median of industry expectations of 210 million units (Table II-1). The key reasons why the industry’s expectations are unreasonably high, in our opinion, are as follows: Chinese demand for new smartphones will likely stay weak (Chart II-2). The mainland smartphone market has become extremely saturated, with 1.3 billion units having been sold in just the past three years – nearly equaling the entire Chinese population. Chinese official data show that each Chinese household owned 2.5 phones on average in 2018, and that the average household size was about three persons (Chart II-3). This suggests that going forward nearly all potential phone demand in China is for replacement phones, and that there is no urgent need for households to buy new phones. Chart II-2Chinese Smartphone Demand: Further Decline In 2020 Chart II-3Chinese Households: No Urgent Need For A New Phone   The Chinese government’s boost to 5G infrastructure investment will likely increase annual installed 5G base stations from 130,000 units last year to about 600,000 to 800,000 this year. However, the total number of 5G base stations will still only account for about 7-9% of total base stations in China in 2020. Hence, geographical coverage will not be sufficiently wide enough to warrant a very high rate of 5G smartphone adoption and penetration. From Chinese consumers’ perspectives, a 5G phone in 2020 will be a ‘nice-to-have,’ but not a ‘must-have.’ Given increasing economic uncertainty and many concerns related to the use of 5G phones, mainland consumers may delay their purchases into 2021 when 5G phone networks will have more geographic coverage.  The number of 5G phone models on the market is expanding, but not that quickly. Consumers may take their time to wait for more models to hit the market before making a 5G phone purchase. For example, Apple will release four 5G phone models, but only in September 2020. Moreover, the price competition between 5G and 4G phones is getting increasingly intense. Smartphone producers have already started to cut prices of their 4G phones aggressively. For example, the price of Apple’s iPhone XS, released in September 2018, has already dropped by about 50% in China. Outside of China, 5G infrastructure development will be much slower. The majority of developed countries will likely give in to pressure from the US and limit their use of Huawei 5G equipment. This will delay infrastructure installation and adoption of 5G throughout the rest of the world because Huawei has the leading and cheapest 5G technology. In 2019, China accounted for about 70% of worldwide 5G smartphone shipments. We reckon that in 2020 Chinese 5G smartphone shipments will be between 120 million and 130 million units. Assuming this accounts for about 70-75% of the world shipment of 5G phones this year, we arrive at our estimate of global 5G smartphone shipments of between 160 million and 180 million units. We agree that 5G technology is revolutionary. Nevertheless, we still believe global semi share prices are presently overhyped by unreasonably optimistic 2020 projections. Overall, investors are pricing global semi stocks using the pace and trajectory of 4G smartphones adoption. However, in 2020 the number and speed of 5G phone penetration will continue lagging that of 4G ones when the latter were introduced in December 2013 (Chart II-4). We agree that 5G technology is revolutionary, and its adoption and penetration will surge in the coming years. Nevertheless, we still believe global semi share prices are presently overhyped by unreasonably optimistic 2020 projections (Chart II-5).  Chart II-4China 5G-Adoption Pace: Slower Than The Case With 4G Chart II-5Net Earnings Of Global Semi Sector: Too Optimistic?   Investment Implications Global semi stocks’ valuations are very elevated, as shown in Chart II-6 and Chart II-7. Besides, semi stocks are overbought, suggesting they could correct meaningfully if lofty growth expectations currently baked into their prices do not materialize in the first half of this year. Chart II-6Global Semi Stocks Valuations: Very Elevated Chart II-7Global Semi Stocks’ Valuations: Very Elevated   The coronavirus outbreak and the resulting delay in 5G phone sales in China in the first half of 2020, along with US pressure on global semi producers not to sell to Huawei, will likely trigger a pullback in semiconductor equities. We recommend patiently waiting for a better entry point for absolute return investors. Within the EM equity universe, we have not been underweight Asian semi stocks because of our negative outlook for the overall EM equity benchmark. The Argentine government will drag out foreign debt negotiations with the IMF and foreign private creditors to secure a more favorable settlement. We remain neutral on Taiwan and overweight Korea. The reason is that DRAM makers such as Samsung and Hynix have rallied much less than TSMC. Besides, geopolitical risks in relation to Taiwan in general and TSMC in particular are rising, warranting a more defensive stance on Taiwanese stocks relative to Korean equities. Ellen JingYuan He Associate Vice President ellenj@bcaresearch.com Argentina’s Eternal Tango With Foreign Creditors Chart III-1Downside Risks To Bond Prices Our view remains that debt negotiations will be drawn-out because the Argentine government is both unwilling and lacks the financial capacity to service public foreign debt. The administration’s recent attitude toward foreign creditors and the IMF have startled markets: sovereign Eurobond bond prices have tanked (Chart III-1). The reasons why the Fernandez administration will play tough ball with creditors and the IMF are as follows: The country’s foreign funding and the public sector debt situations are precarious. Hence, the lower the recovery rate they negotiate with creditors, the more funds will be available to expand social programs and secure domestic political support. Given Fernandez’s and Peronist’s voter base, the government is inclined to please the population at expense of foreign creditors. Moreover, Alberto Fernandez is facing increasing scrutiny from radical Peronists, who want to dissolve the debt altogether. Vice-president Fernandez de Kirchner stated that Argentina should not pay international agents until the economy escapes a recession. To further add to creditors’ frustration, the government has yet to announce a comprehensive economic plan to revive the economy and service outstanding debt. The public foreign currency debt burden is unsustainable – its level stands at $250 billion, about 4 times larger than exports. The country is still in a recession, and economic indicators do not show much improvement. Committing to fiscal austerity to service foreign debt would entail further economic suffering for Argentine businesses and households, something Fernandez rejected throughout his campaign. The authorities are singularly focused on reviving the economy: government expenditures have grown by over 50% annually under the current administration (Chart III-2). Crucially, Argentina has already achieved a large trade surplus and its current account balance is approaching zero (Chart III-3). Assuming exports stay flat, the economy can afford to maintain its current level of imports. This makes the authorities less willing to compromise and more inclined to adopt a tough stance in debt negotiations. Chart III-2Peronist Government Has Again Boosted Fiscal Spending Chart III-3Argentina: Current Account Is Almost Balanced   The risk of this negotiation strategy is that the nation will not be able to raise foreign funding for a while. Nevertheless, the country is currently de facto not receiving any external financing. Hence, this risk is less pressing. Moreover, the administration has already delayed all US$ bond payments until August. This allows them to extend negotiations with creditors over the next six months, thereby increasing uncertainty and further pushing down bond prices. A lower market price on Argentine bonds is beneficial for the government’s negotiation strategy as it implies lower expectations for foreign creditors. Thus, the Fernandez administration’s strategy will be to play hardball and draw-out negotiations as long as possible. We expect Argentina to reach a settlement with creditors no earlier than in the third quarter of this year and at recovery rates below current prices of the nation’s Eurobonds. Russian financial assets will be supported due to improving public sector governance, accelerating domestic demand growth and healthy macro fundamentals. Bottom Line: The government will drag out foreign debt negotiations with the IMF and foreign private creditors to secure a more favorable settlement. Continue to underweight Argentine financial assets over the next several months. Juan Egaña Research Associate juane@bcaresearch.com Russia: Harvesting The Benefits Of Macro Orthodoxy Russian financial markets have shown resilience in face of falling oil prices. This has been the upshot of the nation’s prudent macro policies in recent years. We have been positive on Russia and overweight Russian markets over the past two years and this stance remains intact. Going forward, Russian financial assets will be supported due to improving public sector governance, accelerating domestic demand growth and healthy macro fundamentals: Fiscal policy will be relaxed substantially – both infrastructure and social spending will rise. Specifically, the Kremlin is eager to ramp up the national projects program. This is bullish for domestic demand. Russia’s public finances are currently in a very healthy state. Public debt (14% of GDP) is minimal and foreign public debt (4% of GDP) is tiny. The overall fiscal balance is in large surplus (2.7% of GDP). The current account is also in surplus. Hence, a major boost in fiscal spending will not undermine Russia’s macro stability for some time. As a major sign of policy change, President Putin has sidelined or reduced the authority of policymakers who have been advocating tight fiscal policy. This policy change has been overdue as fiscal policy has been unreasonably tight for longer than required (Chart IV-1). Chart IV-1Russia: Government Spending Has Been Extremely Weak Importantly, the recent changes at the highest levels of government are also positive for governance and productivity. The new Prime Minister Mishustin has earned this appointment for his achievements as the head of the federal tax authority. He has restructured and reorganized the tax department in a way that has boosted its efficiency/productivity substantially and increased tax collection. By promoting him to the head of government, Putin has boosted Mishustin’s authority to reform the entire federal governance system. Given his record of accomplishment, odds are that the new prime minister will succeed in implementing some reforms and restructuring. Thereby, productivity growth that has been stagnant in Russia for a decade could revive modestly. Also, Putin was reluctant to boost infrastructure spending as he was afraid of money being misappropriated without a proper monitoring system. Putin now hopes Mishustin can introduce an efficient governance system of fiscal spending to assure infrastructure projects can be realized with reasonably minimal losses. As to monetary policy, real interest rates are still very high. The prime lending rate is 10%, the policy rate is 6% and nominal GDP growth is 3.3% (Chart IV-2). Weak growth (Chart IV-3) and low inflation will encourage the central bank to continue cutting interest rates. Chart IV-2Russia: Interest Rates Remain Excessively High Chart IV-3Russia's Growth Is Very Sluggish   Finally, the economy does not have any structural excesses and imbalances. The central bank has done a good job in cleansing the banking system and the latter is in healthy shape. Bottom Line: The ruble will be supported by improving productivity, cyclical growth acceleration and a healthy fiscal position. We continue recommending overweighting Russian stocks, local currency bonds and sovereign credit relative to their respective EM benchmarks. Last week, we also recommended a new trade: Short Turkish bank stocks / long Russian bank stocks. The main risk to the absolute performance of Russian markets is another plunge in oil prices and a broad selloff in EM. On November 14, 2019 we recommended absolute return investors to go long Russian local currency bonds and short oil. This strategy remains intact. Finally, we have been recommending the long ruble / short Colombian peso trade since May 31, 2018. This position has generated large gains and we are reiterating it. Arthur Budaghyan Chief Emerging Markets Strategist arthurb@bcaresearch.com Footnotes Equities Recommendations Currencies, Credit And Fixed-Income Recommendations

Highlights Duration: Bond yields will stay low until the daily number of new COVID-19 cases falls to zero, at which point a sell-off is likely. We therefore recommend maintaining below-benchmark portfolio duration on a 6-12 month horizon. Rising odds of a Bernie Sanders presidential win could prevent bond yields from rising at all this year. We may adjust our recommendations in the coming months if this risk increases. Spread Product: Investors should maintain an overweight allocation to spread product versus Treasuries, with a preference for high-yield. Accommodative monetary conditions will ensure that the supply of credit remains ample for some time yet. This will keep defaults low and spreads tight. Monetary Policy: The Fed is in no rush to tighten policy, but has also set a high bar for further cuts. Investors should short August 2020 fed funds futures.  Yields Will Move Higher … But Not Yet Chart 1A Peak In New Cases? Uncertainty about the economic impact of the coronavirus – now officially called COVID-19 – is the cloud that continues to hang over financial markets. Last week, bond yields fell when a change in the definition of what constitutes a confirmed infection caused the number of reported cases to spike. However, even after revisions, the daily number of new cases looks like it may have peaked (Chart 1). The end result is that the 10-year Treasury yield sits at 1.58%, not far from where it was last week (Chart 2). Notably, the 10-year yield continues to shrug off the notable improvement in US economic data (Chart 2, bottom panel), taking its cues instead from COVID-19 headline risk. Even if the downtrend in new COVID-19 cases continues, it is too soon to be looking for higher bond yields. For one thing, the most up-to-date economic data releases were collected during January, before the outbreak. Weaker readings during the next 1-2 months are assured, and investors may not look through the weakness given that many were already skeptical about the prospects for global economic recovery. Our read of the data is that global growth was in the process of bottoming when COVID-19 struck. We therefore expect global growth to move higher once the virus’ impact abates. In terms of timing, using the 2003 SARS outbreak as a comparable, we expect bonds to remain bid until the daily number of new cases falls to zero, at which point a sell-off is likely. Yields continue to shrug off improvements in economic data. It’s not just the long-end of the curve that has responded to COVID-19. The front-end has also moved to price-in high odds of a rate cut in the coming months. Specifically, the overnight index swap curve is priced for a 42 bps decline in the fed funds rate during the next 12 months (Chart 2, panel 2), and the fed funds futures market is pricing a 74% chance of a rate cut by the end of the summer. As we discussed last week, given that any economic impact from COVID-19 will be temporary, we think the bar for a Fed rate cut this year is quite high.1 As such, our Golden Rule of Bond Investing dictates that investors should keep portfolio duration low on a 12-month horizon.2 We also recommend shorting August 2020 fed funds futures, a trade that will earn 23 bps of unlevered return if the Fed stands pat between now and August (Chart 2, panel 3). Turning to corporate credit, we see that, so far, COVID-19’s impact on spreads has been minor. The investment grade corporate bond index spread is only 3 bps wider than at the start of the year, and the junk index spread is only 8 bps wider (Chart 3). Value remains stretched in the investment grade space, but high-yield spreads look quite attractive. The sell-off in the energy sector has boosted the high-yield index spread considerably (Chart 3, bottom 2 panels). We view this as a medium-term buying opportunity for junk. Once the COVID outbreak abates and global growth ticks higher, the oil price is bound to increase, leading to some tightening in energy spreads. Chart 2Bond Yields Driven By COVID Chart 3HY More Attractive Than IG Will Bonds Feel The Bern? Beyond COVID-19, there is one more risk on the horizon this year. Specifically, the risk that Bernie Sanders is elected President in November. This outcome is far from certain. Sanders is currently leading all other candidates in the Democratic Primary, but fivethirtyeight.com’s model puts the odds of a brokered convention at 38%.3 This means that the race is still wide open and might only be settled at the convention in July. But given Sanders’ lead, it is worth considering the bond market implications if he were to become the next President. The most obvious implication is that risk assets (equities and corporate spreads) would respond to Sanders’ agenda of wealth redistribution by selling off. This could spur a flight-to-quality into government bonds, causing Treasury yields to fall. However, that flight-to-quality won’t occur if markets also start to price-in the long-run implications of Sanders’ agenda. I.e. the fact that the redistribution of wealth from capital to labor would lower the economy’s marginal propensity to save, and likely raise inflation expectations, leading to higher interest rates. It’s important to note that there are a lot of hurdles to overcome before Sanders’ full policy agenda is implemented. First he must secure the Democratic nomination, then defeat Donald Trump in the general election. Even after that, he will still need to convince the House and Senate to pass non-watered down versions of his proposals. With such a long road ahead, we don’t think Sanders’ momentum will push bond yields higher in 2020. Rather, the risk is that Sanders’ rise keeps bond yields low in 2020 as risk assets sell off. If Bernie Sanders looks poised to win the nomination, we will consider reducing our 6-12 month allocation to spread product and increasing our recommended portfolio duration. The outlook for the Democratic Primary should become clearer after Super Tuesday on March 3. If Sanders looks poised to win the nomination we will consider reducing our recommended 6-12 month allocation to spread product and increasing our recommended portfolio duration. Bottom Line: Bond yields will stay low until the daily number of new COVID-19 cases falls to zero, at which point a sell-off is likely. We therefore recommend maintaining below-benchmark portfolio duration on a 6-12 month horizon. Rising odds of a Bernie Sanders presidential win could prevent bond yields from rising at all this year. We may adjust our recommendations in the coming months if this risk increases. Investors should maintain an overweight allocation to spread product versus Treasuries, with a preference for junk. Though the credit cycle is far from over (see next section), we may reduce our recommended allocation to spread product versus Treasuries if Sanders’ election chances rise.  Bank Lending Standards Won’t Push Credit Spreads Wider In 2020 The net change in commercial & industrial (C&I) bank lending standards, as reported in the Fed’s quarterly Senior Loan Officer Survey, is a vitally important indicator for the credit cycle. Easing lending standards tend to coincide with a low default rate and falling credit spreads, while tightening lending standards usually coincide with spread widening and a rising default rate. With that in mind, it is mildly concerning that bank lending standards have been fluctuating around neutral levels for quite some time, and have in fact tightened in two of the past five quarters (Chart 4). In this week’s report we consider whether tighter bank lending standards could pose a risk to our overweight spread product view in 2020. Chart 4Bank Lending Standards And Monetary Variables Bank lending standards are such an important credit cycle variable because they tell us about the supply of credit. A corporate default only occurs when credit supply is lower than the amount required for that firm’s survival. On a macro scale, we can think of two main reasons why lenders might restrict the credit supply: They perceive the monetary environment as restrictive. That is, they worry about higher interest rates and slower growth in the future. They perceive corporate balance sheets as being in poor health. That is, they worry that firms won’t be sufficiently profitable to make good on their debts. We find that monetary indicators do a very good job of predicting when lending standards will tighten. Looking back at the past two cycles, lending standards didn’t tighten until after: The yield curve inverted (Chart 4, panel 2). The real fed funds rate was above its estimated equilibrium level (Chart 4, panel 3). Inflation expectations were at or above target levels (Chart 4, bottom panel). Presently, all three of these monetary indicators are supportive. Some portions of the yield curve have been inverted at various times during the past year. But in general, the inversion signal from the yield curve has not been as strong as it was when lending standards tightened in prior cycles. For instance, the 3-year/10-year Treasury slope has not inverted this cycle, and it currently sits at +20 bps (Chart 4, panel 2). Further, the real fed funds rate is below most estimates of its neutral level and the Fed is signaling that it will keep it there for a long time yet. This dovish posture is justified by inflation expectations that remain well below target. It is conceivable that, despite the accommodative monetary environment, banks might be so concerned about poor balance sheet health that they are becoming more cautious with their lending. However, a survey of corporate health metrics doesn’t point to an imminent tightening of bank lending standards either (Chart 5). Chart 5Bank Lending Standards And Corporate Balance Sheet Variables In past cycles, tighter bank lending standards were preceded by: A trough in gross leverage (pre-tax profits over total debt) (Chart 5, panel 2). A peak in interest coverage (Chart 5, panel 3). Negative pre-tax profit growth (Chart 5, panel 4). A peak in profit margins (Chart 5, bottom panel). Currently, gross leverage is the only one of the above four variables that is clearly sending a negative signal. As for the other three, interest coverage and profit margins are barely off their cyclical highs, and profit growth has been fluctuating around zero for three years. If global growth rebounds during the next 12 months, as we expect, then profit growth will also move modestly higher. Bottom Line: Neither monetary nor balance sheet variables point to an imminent tightening of bank lending standards. We expect that the supply of credit will remain ample in 2020, keeping the default rate low and credit spreads tight. A Note On Falling C&I Loan Demand In addition to questions about lending standards, the Fed’s Senior Loan Officer Survey also asks banks to report whether they are seeing stronger or weaker demand for C&I loans. In response, banks have reported weaker C&I loan demand for six consecutive quarters, ending in Q4 2019. Historically, it is unusual for C&I loan demand to fall without a concurrent tightening in lending standards (Chart 6). Chart 6Explaining Weakening Loan Demand We also see the impact of weaker loan demand in the hard data. C&I loan growth has been falling since early 2019 (Chart 6, panel 2) and net corporate bond issuance had been on a sharp downtrend since 2015, before moving higher last year (Chart 6, bottom panel). So what’s going on with C&I loan demand? We can think of two reasons why firms might seek out less credit. First, they may face a dearth of investment opportunities, or alternatively, they might perceive some benefit from carrying less debt on their balance sheets. On the first point, we find that new orders for core capital goods do a very good job explaining the swings in C&I lending (Chart 7). Specifically, we see that the global growth slowdown of 2015/16 drove both investment spending and C&I lending lower. Then, both series recovered in 2017/18 before moving down again during last year’s slowdown. Surveys about firms’ capital spending plans also dropped last year, consistent with the deceleration in C&I lending, but remain at high levels (Chart 7, bottom three panels). All of this suggests that C&I loan growth will recover this year as global growth improves and the investment landscape brightens. Capital goods new orders do a good job explaining C&I lending. Corporate bond issuance has followed a different path from C&I lending during the past few years. Specifically, bond issuance slowed in 2015/16 as investment spending dried up. But it did not recover in 2017/18 the way that investment spending and C&I lending did. This appears to be a result of the 2018 corporate tax cuts and repatriation holiday. Chart 8 shows that the Financing Gap – the difference between capex spending and retained earnings – plunged in 2018 because firms suddenly received a huge influx of retained earnings. The influx came in part from the lower tax rate, but mostly from repatriated cash that had been stranded overseas. Simply, firms didn’t need to issue bonds to finance their investment plans in 2018 because they had a lot more cash on hand. Chart 7C&I Lending Follows ##br##Investment Chart 8A Negative Financing Gap Limits The Need For Debt What about the possibility that firms are demanding less debt because they are trying to clean up their balance sheets? Beyond a few anecdotes, we don’t see much support for this idea. In fact, an equity index of firms with low debt/asset ratios has been underperforming an index of firms with high debt/asset ratios (Chart 9). This suggests that there is currently little reward for firms that are paying down debt. Chart 9Firms Not Rewarded For Healthy Balance Sheets Bottom Line: Weaker demand for C&I loans is a result of the recent global growth downturn and decline in investment spending. It is not a harbinger of the end of the credit cycle. Loan demand should improve as global growth rebounds this year. Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see US Bond Strategy Weekly Report, “How Are Inflation Expectations Adapting?”, dated February 11, 2020, available at usbs.bcaresearch.com 2 For further details on our Golden Rule of Bond Investing please see US Bond Strategy Special Report, “The Golden Rule of Bond Investing”, dated July 24, 2018, available at usbs.bcaresearch.com 3 https://projects.fivethirtyeight.com/2020-primary-forecast/?ex_cid=rrpromo Fixed Income Sector Performance Recommended Portfolio Specification

Highlights Provided that the coronavirus outbreak is contained, global growth should accelerate over the course of 2020. Stocks usually rise when the economy is strengthening. But could this time be different? We explore five scenarios in which the stock market could decouple from the economy: 1) The economy holds up, but stretched valuations bring down equities, especially high-flying growth stocks; 2) Bond yields rise in response to faster growth, hurting equities in the process; 3) A strong US economy lifts the value of the dollar, denting multinational profits and tightening financial conditions abroad; 4) Faster wage growth cuts into corporate profits; and 5) Redistributionist politicians seek to shift income from capital to labor. We are not too concerned about the first four scenarios, but we do worry about the fifth, especially now that betting markets are giving Bernie Sanders a nearly 50% chance of becoming the Democratic nominee. Matters should be clearer by mid-March, by which time more than 60% of Democratic delegates will have been awarded. If Bernie Sanders does emerge as the nominee at that point, we will consider trimming back our bullish cyclical bias towards stocks. Coronavirus: A Break In The Clouds? Chart 1Coronavirus Remains Mostly Contained To China Investors continue to grapple with two distinct narratives about how the coronavirus outbreak is unfolding. On the pessimistic side, some contend that the true number of infections in China is much higher than the Chinese authorities are disclosing. How else, they ask, can one explain why the government has taken the extreme step of imposing some form of quarantine on 400 million of its own people? More optimistic observers argue that the Chinese government is simply being proactive. While the number of cases in Hubei province spiked yesterday, this was due to a loosening in the definition for what constitutes a confirmed infection. Whereas previously a positive laboratory test was required, now a positive imaging-based clinical examination will suffice. Under the new definition, the number of newly confirmed cases fell from 6,528 on February 11th to 4,273 on February 12th. Under the old definition, newly diagnosed cases peaked on February 2nd (Chart 1). The revised definition adopted in Hubei brought the mortality rate in the province down to 2.7%. The mortality rate observed in the rest of China is 0.5%. The share of all cases in China originating in Hubei also rose to 81%. Even before the rule change, the share of cases diagnosed in Hubei had risen from 52% on January 26th to 75% on February 11th. This suggests progress in limiting the outbreak to the province. Critically, the number of cases in the rest of the world remains low. In the US, a total of 13 cases have been confirmed as of February 12th, just two more than the 11 reported on February 2nd. The Exception To The Rule? Provided that the coronavirus outbreak is contained, global growth should bounce back forcefully in the second quarter. If that were to occur, history suggests that equities will continue to rally, while bond prices will fall (Chart 2). But could history fail to repeat itself? In this week’s report, we explore five scenarios in which that may happen. Scenario 1: Stretched valuations bring down equities, especially high-flying growth stocks Stocks have moved up considerably since their December 2018 lows. This suggests that investors have become more confident about the economic outlook. Nevertheless, while most investors may no longer be worried about an imminent recession, they do not foresee a sharp acceleration in global growth either. This is evidenced by the fact that cyclical stocks have generally underperformed defensives (Chart 3). Oil prices have also languished, while copper prices are back near a 2.5-year low (Chart 4). Chart 2Stocks Usually Outperform Bonds When Global Growth Is Accelerating Chart 3Cyclicals Have Failed To Outperform Defensives   At the broad index level, global equities trade at 16.7-times forward earnings. Conceptually, the inverse of the PE ratio – the earnings yield – should serve as a reasonable guide for the total real return that equities will deliver over the long haul.1 At 6%, the global earnings yield still points to decent returns for global stocks. Relative to bonds, the case for owning stocks is even more compelling. The equity risk premium, which one can compute as the earnings yield minus the real bond yield, remains well above its historic average (Chart 5). Chart 4Commodity Prices Have Taken It On The Chin Chart 5Relative Valuations Favor Equities   That said, there are pockets where valuations have gotten stretched. US equities trade at 19.5-times forward earnings compared to 14.1-times in the rest of the world. Growth stocks, in particular, have gotten very expensive (Chart 6). The five largest stocks in the S&P 500 (Apple, Microsoft, Amazon, Alphabet, and Facebook) now account for 18% of the index, the same share that the top five stocks (Microsoft, Cisco, GE, Intel, and Exxon) commanded in 2000. The big risk for stocks is that wages go up not because the overall size of the economic pie is growing, but because policies are implemented that shift a bigger share of the pie from capital to labor. Despite the similarities between today and the dotcom era, there are a few critical differences – most of which make us less worried about the current state of affairs. First, while tech valuations are currently stretched, they are not in bubble territory. The NASDAQ Composite trades at 30-times trailing earnings. At its peak in March 2000, the tech-heavy index traded at more than 70-times earnings (Chart 7). Chart 6Growth Stocks Have Become Expensive Relative To Value Stocks Chart 7Not Yet Partying Like 1999   Second, IPO activity has also been more muted today than during the dotcom boom (Chart 8). Only 110 companies went public last year, with the gain on the first day of trading averaging 24%. In 1999, 476 companies went public. The average first day gain was 71%. Meanwhile, companies continue to buy up their shares. The buyback yield stands at 3%, twice as high as in the late 1990s. Third, there is no capex overhang like in the late 1990s (Chart 9). This reduces the odds of a 2001-recession scenario where falling equity prices prompted companies to pare back capital expenditures, leading to rising unemployment and even lower equity prices. Chart 8IPO Activity Is Muted Today Compared To The Late 1990s Chart 9No Capex Boom This Time   Scenario 2: Bond yields rise in response to faster growth, hurting equities in the process The period between November 2018 and September 2019 was an odd one for the stock-to-bond correlation. If one looks at daily data, stocks did best when bond yields were rising. Yet, for the period as a whole, stocks finished higher while bond yields finished lower (Chart 10). Chart 10Daily Changes: S&P 500 Vs. 10-Year Treasury Yield How can one explain this seeming paradox? The answer is that the underlying trend in bond yields was squarely to the downside last year. While yields did rise modestly on days when equities rallied, yields fell sharply on days when equities swooned. If one zooms out, one sees the underlying trend, whereas if one zooms in, one only sees the wiggles around the trend. Bond yields trended lower last year because the Fed and most other central banks were delivering one dose of dovish medicine after another. This year, however, the Fed is on hold, and while a few central banks may still cut rates, global monetary policy is unlikely to become much looser. This means that bond yields are likely to drift higher if economic growth surprises on the upside. Will rising bond yields sabotage the stock market? We do not think so. Stocks crashed in late 2018 because investors became convinced that US monetary policy had turned restrictive after the Fed had raised rates by a cumulative 200 basis points over the prior two years. The fact that the Laubach-Williams model, one of the most widely followed models of the neutral rate, showed that real rates had moved above their equilibrium level did not help sentiment (Chart 11). Chart 11The Fed Will Keep Policy Easy For The Time Being Chart 12Stocks Do Well When Earnings And Growth Surprise On The Upside Today, real rates are about 100 basis points below the Laubach-Williams estimate. This will not change anytime soon, given that the Fed is likely to remain on hold at least until the end of the year. So long as rates stay put, monetary policy will remain accommodative, allowing the economy to grow at a solid pace. Granted, rising long-term bond yields will reduce the present value of future cash flows, thus potentially hurting stocks. However, as we discussed three weeks ago, the discount rate is not the only thing that affects equity valuations.2 The expected growth rate of earnings matters too. As Chart 12 shows, global equity returns are highly sensitive to earning revisions. While earnings may disappoint in the first quarter due to the economic damage from the coronavirus, they should bounce back during the remainder of this year. This should pave the way for higher equity prices. Scenario 3: A strong US economy lifts the value of the dollar, denting multinational profits and tightening financial conditions abroad The US is a fairly closed economy. Imports and exports account for only 14.6% and 11.7% of GDP, respectively. In contrast, the US stock market is very exposed to the rest of the world. S&P 500 companies derive over 40% of their sales from abroad. As such, changes in the value of the dollar tend to have a bigger impact on Wall Street than on Main Street. Estimating the degree to which a stronger dollar reduces S&P 500 profits is no easy task. Direct estimates that measure the currency translation effect on overseas profits from a stronger dollar tend to yield fairly modest results, typically showing that a 10% appreciation in the trade-weighted dollar reduces S&P 500 profits by about 2%. These estimates, however, generally do not take into account feedback loops between a strengthening dollar and global financial conditions (Chart 13). According to the Bank of International Settlements, $12 trillion of dollar-denominated debt has been issued outside the US. A stronger dollar makes it more challenging to service this debt, which can put a significant strain on borrowers. As a result, a vicious cycle can erupt where a stronger dollar leads to tighter financial conditions, which in turn lead to weaker global growth and an even stronger dollar. Chart 13A Strong US Dollar Could Tighten Global Financial Conditions, Leading To Lower Equity Prices, Especially In EM Such an outcome cannot be dismissed, especially if the spread of the coronavirus fuels significant foreign inflows into the safe-haven US Treasury market. Nevertheless, we continue to see it as a low-probability event given the tailwinds to global growth, including the lagged effects of last year’s decline in bond yields, an improvement in the global manufacturing inventory cycle, diminished Brexit and trade war risks, and ongoing policy stimulus out of China. In fact, one can more easily envision the opposite outcome – a virtuous cycle of dollar weakness, leading to easier global financial conditions, stronger growth, and ultimately, an even weaker dollar (Chart 14). In such an environment, earnings growth is likely to accelerate (Chart 15). Chart 14The Dollar Is A Countercyclical Currency Chart 15The Virtuous Cycle Of Dollar Easing     Scenario 4: Faster wage growth cuts into corporate profits Labor compensation is the largest expense for most companies. Thus, it stands to reason that faster wage growth could depress earnings, and by extension, share prices. Although this is possible conceptually, in practice, it happens less often than one might guess. Chart 16 shows that rising wage growth is positively correlated with earnings. The bottom panel of the chart explains why: Wages tend to rise most quickly when sales are growing rapidly. Strong demand growth adds to revenues, while allowing companies to spread fixed costs over a large amount of output. The resulting improvement in “operating leverage” helps buffer profit margins from higher wages. Scenario 5: Redistributionist politicians seek to shift income from capital to labor As long as wages are rising against a backdrop of fast sales growth, equities will fare well. The big risk for stocks is that wages go up not because the overall size of the economic pie is growing, but because policies are implemented that shift a bigger share of the pie from capital to labor. Bernie Sanders has promised to do just that. The S&P 500 has tended to increase when Sanders’ perceived chances of winning the Democrat nomination have risen (Chart 17). Investors have apparently concluded that Trump would clobber Sanders in a presidential race. Hence, the better Sanders performs in the primaries, the more likely Trump is to be re-elected. Chart 16Stocks Tend To Do Best When Wage Growth Is Rising Chart 17The Sanders Effect On Stocks   Is this really a safe assumption? We are not so sure. Sanders has still beaten Trump in 49 of the last 54 head-to-head polls tracked by Realclearpolitics over the past 12 months. Sanders tends to appeal to white working class voters – the same demographic that propelled Trump into office. Sanders is also benefiting from a secular leftward shift in voter attitudes on economic issues. According to a recent Gallup poll, 47% of Americans believe that governments should do more to solve problems, up from 36% in 2010. Almost 40% of Americans have a positive view on socialism (Chart 18). Today’s youth in particular is enamored with left-wing ideology (Chart 19). Chart 18The US Is Moving To The Left Chart 19Woke Millennials Cozying Up To Socialism It’s not just the Democratic voters who are trending left. Some prominent Republicans are having second thoughts too. Tucker Carlson is probably the best leading indicator for where the Republican Party is heading. His attacks on “woke capitalism” have become a staple of his popular evening show.3 It is not surprising why many Republicans are having a change of heart. For decades, the Republican Party has been a cheap date for corporate interests: It has given businesses what they want – lower taxes, less regulation, etc. – without asking for much in return (aside from campaign contributions, of course). This has allowed corporations to focus on appealing to left-wing interests by taking increasingly strident positions on a variety of social issues. The fact that some of these positions – such as support for open-border immigration policies – are a boon for profits has only increased their appeal. The risk for corporations is that they end up with no real political support. If the Democrats move further to the left, “soak the rich” policies will become popular no matter how much virtue signaling corporate leaders deliver. Likewise, if Republicans abandon big businesses, today’s fat profit margins will become a thing of the past. When The Music Ends The current market climate resembles a Parisian ball on the eve of the French Revolution. The music is still playing, but the discontent among the commoners outside is growing. The question is when will this discontent boil over? Trump’s victory in 2016 represented a shot across the bow of the political establishment. Fortunately for corporate interests, aside from his protectionist impulses, Trump has been on their side. Bernie Sanders would not be so friendly. Matters should be clearer by mid-March. Super Tuesday takes place on March 3rd. By March 17th, more than 60% of Democratic delegates will have been awarded. If Bernie Sanders emerges as the likely nominee at that point, we will consider trimming back our bullish cyclical 12-month bias towards stocks. Peter Berezin Chief Global Strategist peterb@bcaresearch.com   Footnotes 1  Please see Global Investment Strategy Special Report, “TINA To The Rescue?” dated August 23, 2019. 2  Please see Global investment Strategy Weekly Report, “Bond Yields: How High Is Too High?” dated January 17, 2020. 3  Ian Schwartz, “Tucker Carlson: Elizabeth Warren's "Economic Patriotism" Plan "Sounds Like Donald Trump At His Best," realclearpolitics, June 6, 2019. Global Investment Strategy View Matrix MacroQuant Model And Current Subjective Scores Strategic Recommendations Closed Trades

Highlights Chart 1The 2003 SARS Roadmap The bond market impact from the coronavirus has already been substantial. The 10-year Treasury yield has fallen back to 1.51%, below the fed funds rate. Meanwhile, the investment grade corporate bond index spread is back above 100 bps, from a January low of 93 bps. The 2003 SARS crisis is the best roadmap we can apply to the current situation. Back then, Treasury yields also fell sharply but then rebounded just as quickly when the number of SARS cases peaked (Chart 1). The impact on corporate bond excess returns was more short-lived (Chart 1, bottom panel). Like in 2003, we expect that bond yields will rise once the number of coronavirus cases peaks, but it is difficult to put a timeframe on how long that will take. The economic impact from the virus could also weigh on global PMI surveys during the next few months, delaying the move higher in Treasury yields we anticipated earlier this year. In short, we continue to expect higher bond yields and tighter credit spreads in 2020, but those moves will be delayed until markets are confident that the virus has stopped spreading. Feature Investment Grade: Neutral Chart 2Investment Grade Market Overview Investment grade corporate bonds underperformed the duration-equivalent Treasury index by 80 basis points in January. The sector actually outpaced the Treasury benchmark by 7 bps until January 21 when the impact of the coronavirus started to push spreads wider. As stated on page 1, we expect the impact of the coronavirus on corporate spreads to be short lived. Beyond that, low inflation expectations will keep monetary conditions accommodative. This in turn will encourage banks to ease credit supply, keeping defaults at bay and providing a strong tailwind for corporate bond returns.1 Yesterday’s Fed Senior Loan Officer survey showed a slight easing of C&I lending standards in Q4 2019, reversing the tightening that occurred in the third quarter (Chart 2). We expect that accommodative Fed policy will lead to continued easing of C&I lending standards for the remainder of the year. Despite the positive tailwind from accommodative Fed policy and easing bank lending standards, investment grade corporate bond spreads are quite expensive. Spreads for all credit tiers are below our targets (panels 2 & 3).2 As a result, we advise only a neutral allocation to investment grade corporate bonds. We also recommend increasing exposure to Agency MBS in place of corporate bonds rated A or higher (see page 7). Table 3ACorporate Sector Relative Valuation And Recommended Allocation* Table 3BCorporate Sector Risk Vs. Reward* High-Yield Overweight Chart 3High-Yield Market Overview High-Yield underperformed the duration-equivalent Treasury index by 111 basis points in January. Junk outperformed the Treasury benchmark by 30 bps until January 21 when the coronavirus outbreak sent spreads sharply wider. Once the negative impact of the coronavirus passes, junk spreads will have plenty of room to tighten in 2020. In fact, the junk index spread is now at 390 bps, 154 bps above our target (Chart 3).3 While spreads for all junk credit tiers are currently above our targets, Caa-rated bonds look particularly cheap. We analyzed the divergence between Caa and the rest of the junk index in a recent report and came to two conclusions.4 First, the historical data show that 12-month periods of overall junk bond outperformance are more likely to be followed by underperformance if Caa is the worst performing credit tier. Second, we can identify several reasons for 2019’s Caa spread widening that make us inclined to downplay any negative signal. Specifically, we note that the Caa credit tier’s exposure to the shale oil sector is responsible for the bulk of 2019’s underperformance (bottom panel). Absent significant further declines in the oil price, this sector now has room to recover.   MBS: Overweight Chart 4MBS Market Overview Mortgage-Backed Securities underperformed the duration-equivalent Treasury index by 53 basis points in January. The sector was only lagging the Treasury benchmark by 7 bps as of January 21, when the coronavirus outbreak sent spreads wider. The conventional 30-year zero-volatility spread widened 8 bps in January, driven by a 7 bps widening of the option-adjusted spread (OAS) and a 1 bp increase in expected prepayment losses (aka option cost). The fact that expected prepayment losses only rose by a single basis point even though the 30-year mortgage rate fell by 23 bps is notable. It speaks to the high level of refi burnout in the mortgage market, which is a key reason why we prefer mortgage-backed securities over investment grade corporate bonds in our portfolio. Essentially, most homeowners have already had at least one opportunity to refinance during the past few years, so prepayment risk is low even if rates fall further. Competitive expected compensation is another reason to move into Agency MBS. The conventional 30-year MBS OAS is 49 bps, only 7 bps below the spread offered by Aa-rated corporate bonds (Chart 4). Also, spreads for all investment grade corporate bond credit tiers are below our cyclical targets. Risk-adjusted compensation favors MBS even more strongly. The Excess Return Bond Map in Appendix C shows that Agency MBS plot well to the right of investment grade corporates. This means that the sector is less likely to see losses versus Treasuries on a 12-month horizon. Government-Related: Underweight Chart 5Government-Related Market Overview The Government-Related index underperformed the duration-equivalent Treasury index by 14 basis points in January. The index was up 2 bps versus the Treasury benchmark until January 21, when the coronavirus outbreak hit. Sovereign debt underperformed duration-equivalent Treasuries by 99 bps on the month, and Foreign Agencies underperformed by 28 bps. Local Authorities, however, bested the Treasury benchmark by 60 bps. Domestic Agency bonds underperformed Treasuries by 2 bps in January, while Supranationals outperformed by 2 bps. We continue to recommend an underweight allocation to USD-denominated sovereign bonds, given that spreads remain expensive compared to US corporate credit (Chart 5). However, we noted in a recent report that Mexican and Saudi Arabian sovereigns look attractive on a risk/reward basis.5 This is also true for Local Authorities and Foreign Agencies, as shown in the Bond Map in Appendix C. Our Emerging Markets Strategy service also thinks that worries about Mexico’s fiscal position are overblown, and that bond yields embed too high of a risk premium (bottom panel).6  Municipal Bonds: Overweight Chart 6Municipal Market Overview Municipal bonds underperformed the duration-equivalent Treasury index by 33 basis points in January (before adjusting for the tax advantage). They were up 39 bps versus the Treasury index before the coronavirus outbreak hit on January 21. The average Aaa-rated Municipal / Treasury (M/T) yield ratio swung around during the month, but settled close to where it began at 77% (Chart 6). We upgraded municipal bonds in early October, as yield ratios had become significantly more attractive, especially at the long-end of the Aaa curve (panel 2).7 Yield ratios have tightened a lot since then, but value remains at long maturities. Specifically, the 2-year, 5-year and 10-year M/T yield ratios are all below average pre-crisis levels at 62%, 65% and 78%, respectively. But 20-year and 30-year yield ratios stand at 89% and 93%, respectively, above average pre-crisis levels. Fundamentally, state and local balance sheets remain solid. Our Municipal Health Monitor is in “improving health” territory and state & local government interest coverage has improved considerably in recent quarters (bottom panel). Both of these trends are consistent with muni ratings upgrades continuing to outpace downgrades going forward. Treasury Curve: Maintain A Barbell Curve Positioning Chart 7Treasury Yield Curve Overview The Treasury curve bull-flattened dramatically in January. Treasury yields declined across the curve, and the 2/10 slope flattened from 34 bps to 18 bps. The 5/30 slope flattened from 70 bps to 67 bps. Despite the significant flattening, the 2/10 slope remains near the middle of our target 0 – 50 bps range for 2020, and we anticipate some bear-steepening once the coronavirus is contained.8 The front-end of the curve also moved in January to price-in 57 bps of Fed rate cuts during the next 12 months (Chart 7). At the beginning of the year the curve was priced for only 14 bps of rate cuts. We expect that the Fed would respond with rate cuts if the coronavirus epidemic worsens, leading to inversion of the 2/10 yield curve. However, for the time being the safer bet is that the virus will be contained relatively quickly and the Fed will remain on hold for all of 2020. Based on this view, we continue to recommend holding a barbelled Treasury portfolio. Specifically, we favor holding a 2/30 barbell versus the 5-year bullet, in duration-matched terms. The position offers positive carry and looks attractive on our yield curve models (see Appendix B).9  TIPS: Overweight Chart 8Inflation Compensation TIPS underperformed the duration-equivalent Treasury index by 75 basis points in January. The 10-year TIPS breakeven inflation rate fell 12 bps on the month and currently sits at 1.66%. The 5-year/5-year forward TIPS breakeven inflation rate fell 16 bps on the month and currently sits at 1.71%. Both rates remain well below the 2.3%-2.5% range consistent with the Fed’s target. The divergence between the actual inflation data and inflation expectations remains stark. Trimmed mean PCE inflation has been fluctuating around the Fed’s target since mid-2018 (Chart 8). However, long-maturity TIPS breakeven inflation rates remain stubbornly low. It takes time for expectations to adapt to a changing macro environment, but even accounting for those long lags, our Adaptive Expectations Model pegs the 10-year TIPS breakeven inflation rate as 31 bps too low (panel 4).10 It is highly likely that the Fed will have to tolerate some overshoot of its 2% inflation target in order to re-anchor long-term inflation expectations. As a result, the actual inflation data will lead expectations higher, causing the TIPS breakeven inflation curve to flatten.11 ABS: Underweight Chart 9ABS Market Overview Asset-Backed Securities outperformed the duration-equivalent Treasury index by 32 basis points in January. The index option-adjusted spread for Aaa-rated ABS tightened 14 bps on the month. It currently sits at 26 bps, below its minimum pre-crisis level (Chart 9). Our Excess Return Bond Map (see Appendix C) shows that Aaa-rated consumer ABS ranks among the most defensive US spread products. This explains why the sector performed so well in January when other spread sectors struggled. ABS also offer higher expected returns than other low-risk sectors such as Domestic Agency bonds and Supranationals. However, we remain wary of allocating too much to consumer ABS because credit trends are slowly shifting in the wrong direction. The consumer credit delinquency rate remains low, but has put in a clear bottom. This is also true for the household interest expense ratio (panel 3). Senior Loan Officers also continue to tighten lending standards for both credit cards and auto loans. Tighter lending standards usually coincide with rising delinquencies (bottom panel). Non-Agency CMBS: Neutral Chart 10CMBS Market Overview Non-Agency Commercial Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 43 basis points in January. The index option-adjusted spread for non-agency CMBS tightened 6 bps on the month. It currently sits at 67 bps, below its average pre-crisis level (Chart 10). In last week’s Special Report, we explored how low interest rates have boosted commercial real estate (CRE) prices this cycle, and concluded that a sharp drawdown in CRE prices is likely only when inflation starts to pick up steam.12 In that report we also mentioned that non-agency Aaa-rated CMBS spreads look attractive relative to US corporate bonds from a risk/reward perspective (see our Excess Return Bond Map in Appendix C), and that the macro environment is only slightly unfavorable for CMBS spreads. Specifically, CRE bank lending standards are just in “net tightening” territory. But both lending standards and loan demand are very close to neutral (bottom 2 panels). Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 34 basis points in January. The index option-adjusted spread tightened 4 bps on the month to reach 54 bps. The Excess Return Bond Map in Appendix C shows that Agency CMBS offer a compelling risk/reward trade-off. An overweight allocation to this sector remains appropriate. Appendix A: The Golden Rule Of Bond Investing We follow a two-step process to formulate recommendations for bond portfolio duration. First, we determine the change in the federal funds rate that is priced into the yield curve for the next 12 months. Second, we decide – based on our assessments of the economy and Fed policy – whether the change in the fed funds rate will exceed or fall short of what is priced into the curve. Most of the time, a correct answer to this question leads to the appropriate duration call. We call this framework the Golden Rule Of Bond Investing, and we demonstrated its effectiveness in the US Bond Strategy Special Report, “The Golden Rule Of Bond Investing”, dated July 24, 2018, available at usbs.bcaresearch.com. Chart 11 illustrates the Golden Rule’s track record by showing that the Bloomberg Barclays Treasury Master Index tends to outperform cash when rate hikes fall short of 12-month expectations, and vice-versa. At present, the market is priced for 57 basis points of cuts during the next 12 months. We anticipate a flat fed funds rate over that time horizon, and therefore anticipate that below-benchmark portfolio duration positions will profit. Chart 11The Golden Rule's Track Record We can also use our Golden Rule framework to make 12-month total return and excess return forecasts for the Bloomberg Barclays Treasury index under different scenarios for the fed funds rate. Excess returns are relative to the Bloomberg Barclays Cash index. To forecast total returns we first calculate the 12-month fed funds rate surprise in each scenario by comparing the assumed change in the fed funds rate to the current value of our 12-month discounter. This rate hike surprise is then mapped to an expected change in the Treasury index yield using a regression based on the historical relationship between those two variables. Finally, we apply the expected change in index yield to the current characteristics (yield, duration and convexity) of the Treasury index to estimate total returns on a 12-month horizon. The below tables present those results, along with 95% confidence intervals. Excess returns are calculated by subtracting assumed cash returns in each scenario from our total return projections. Appendix B: Butterfly Strategy Valuations The following tables present the current read-outs from our butterfly spread models. We use these models to identify opportunities to take duration-neutral positions across the Treasury curve. The following two Special Reports explain the models in more detail: US Bond Strategy Special Report, “Bullets, Barbells And Butterflies”, dated July 25, 2017, available at usbs.bcaresearch.com US Bond Strategy Special Report, “More Bullets, Barbells And Butterflies”, dated May 15, 2018, available at usbs.bcaresearch.com Table 4 shows the raw residuals from each model. A positive value indicates that the bullet is cheap relative to the duration-matched barbell. A negative value indicates that the barbell is cheap relative to the bullet. Table 4Butterfly Strategy Valuation: Raw Residuals In Basis Points (As Of January 31, 2020) Table 5 scales the raw residuals in Table 4 by their historical means and standard deviations. This facilitates comparison between the different butterfly spreads. Table 5Butterfly Strategy Valuation: Standardized Residuals (As Of January 31, 2020) Table 6 flips the models on their heads. It shows the change in the slope between the two barbell maturities that must be realized during the next six months to make returns between the bullet and barbell equal. For example, a reading of 33 bps in the 5 over 2/10 cell means that we would only expect the 5-year to outperform the 2/10 if the 2/10 slope steepens by more than 33 bps during the next six months. Otherwise, we would expect the 2/10 barbell to outperform the 5-year bullet. Table 6Discounted Slope Change During Next 6 Months (BPs) Appendix C: Excess Return Bond Map The Excess Return Bond Map is used to assess the relative risk/reward trade-off between different sectors of the US bond market. It is a purely computational exercise and does not impose any macroeconomic view. The Map’s vertical axis shows 12-month expected excess returns. These are proxied by each sector’s option-adjusted spread. Sectors plotting further toward the top of the Map have higher expected returns and vice-versa. Our novel risk measure called the “Risk Of Losing 100 bps” is shown on the Map’s horizontal axis. To calculate it, we first compute the spread widening required on a 12-month horizon for each sector to lose 100 bps or more relative to a duration-matched position in Treasury securities. Then, we divide that amount of spread widening by each sector’s historical spread volatility. The end result is the number of standard deviations of 12-month spread widening required for each sector to lose 100 bps or more versus a position in Treasuries. Lower risk sectors plot further to the right of the Map, and higher risk sectors plot further to the left. Excess Return Bond Map (As Of January 31, 2020)​​​​​​​ Footnotes 1 Please see US Bond Strategy Special Report, “2020 Key Views: US Fixed Income”, dated December 10, 2019, available at usbs.bcaresearch.com 2  For details on how we calculate our spread targets please see US Bond Strategy Weekly Report, “The Value In Corporate Bonds”, dated February 19, 2019, available at usbs.bcaresearch.com 3  For details on how we calculate our spread targets please see US Bond Strategy Weekly Report, “The Value In Corporate Bonds”, dated February 19, 2019, available at usbs.bcaresearch.com 4  Please see US Bond Strategy Weekly Report, “Caa-Rated Bonds: Warning Sign Or Buying Opportunity?”, dated November 26, 2019, available at usbs.bcaresearch.com 5 Please see US Bond Strategy Weekly Report, “A Perspective On Risk And Reward”, dated October 15, 2019, available at usbs.bcaresearch.com 6 Please see Emerging Markets Strategy Weekly Report, “Country Insights: Malaysia, Mexico & Central Europe”, dated October 31, 2019, available at ems.bcaresearch.com 7 Please see US Bond Strategy Weekly Report, “Two Themes And Two Trades”, dated October 1, 2019, available at usbs.bcaresearch.com 8 Please see US Bond Strategy Special Report, “2020 Key Views: US Fixed Income”, dated December 10, 2019, available at usbs.bcaresearch.com 9 For further details on our recommended yield curve trade please see US Bond Strategy Weekly Report, “The Best Spot On The Yield Curve”, dated January 21, 2020, available at usbs.bcaresearch.com 10 For further details on our Adaptive Expectations Model please see US Bond Strategy Weekly Report, “Adaptive Expectations In The TIPS Market”, dated November 20, 2018, available at usbs.bcaresearch.com 11  Please see US Bond Strategy Special Report, “2020 Key Views: US Fixed Income”, dated December 10, 2019, available at usbs.bcaresearch.com 12  Please see US Investment Strategy / US Bond Strategy Special Report, “Commercial Real Estate And US Financial Stability”, dated January 27, 2020, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification Corporate Sector Relative Valuation And Recommended Allocation

Highlights The intense focus on the weakening of global oil demand expected in the wake of another coronavirus outbreak in China – dubbed 2019-nCoV – obscures likely supply-side responses by OPEC 2.0. The producer coalition likely will rebalance markets by extending production cuts from end-March to at least the end of June when it meets in Vienna March 5-6.  OPEC 2.0 producers will be exquisitely sensitive to Asian refiner demand. They will use it as a gauge for how severe 2019-nCoV’s impact will be on EM demand, and adjust production and exports accordingly. On the demand side, it is difficult to analogue the 2019-nCoV outbreak to the 2003 SARS outbreak, given all the conflicting fundamentals at play at that time.  Forward curves for the principal benchmark crude oils – Brent and WTI – remain backwardated, in spite of the 2019-nCoV-related sell-off. Longer-dated WTI (out to December 2023) traded below $50/bbl earlier in the week, roughly in line with shale-breakeven costs reported by the Dallas Fed earlier this month. This likely will continue to pressure capex in the US shales, keeping future supply growth constrained. Feature Forward curves for the principal benchmark crude oils – Brent and WTI – remain backwardated, in spite of the 2019-nCoV-related sell-off. Chart of the WeekChina's Oil Demand Drives Global Growth Oil markets are rightly focused on the demand implications of the 2019-nCoV outbreak in China.1 Since 2000, China has accounted for 42% of annual oil-demand growth worldwide (Chart of the Week). China is second only to the US in oil demand, accounting for 14% of total global demand of 100.7mm b/d at the end of 2019; its oil imports averaged more than 10mm b/d last year, and are expected to remain strong as it continues to build out its refining sector. Chart 2Asian Air Travel Hit Hard By SARS Historical analogues for 2019-nCoV are difficult. The immediate analogue is the SARS coronavirus outbreak identified in China in February 2003, which lasted six months and hit Asian air travel especially hard (Chart 2). During the height of the SARS outbreak in April 2003, air-travel passenger demand in Asia plunged 45%, according to the International Air Transport Association (IATA). This pushed jet fuel prices lower in Asia and in other key markets, along with distillate prices generally (Chart 3).2 Chart 3Fundamental Supply-Demand Balances Support Higher Crude Oil Prices China now is an extremely large share of global jet fuel consumption. Chart 4BCA Models, Base Metals Prices Suggest SARS Effect Was Short-Lived The industry now is more reliant on Chinese travelers. Since 2003, the number of annual air passengers has more than doubled, with China growing to become the world’s largest outbound travel market. In 2003, close to 7mm passengers from China traveled on international flights. By 2018, that number had grown close to 64mm people, according to China’s aviation authority.  As Chart 2 demonstrates, China now is an extremely large share of global jet fuel consumption. Still, oil is a global market – the avoidance of China during the SARS outbreak in 2003 would have impacted global air travel, and, as a result, global jet-fuel prices. Our proprietary EM commodity-demand models and the behavior of base metals prices, which were and remain heavily influenced by China’s economy, suggest China’s GDP growth slowed in 2003 (mainly 1H03) because of the SARS outbreak (Chart 4).  The LME’s base metals index fell 9% between February and July 2003, while copper prices fell 11%. By year-end, these markets had fully recovered. Oil-Supply Management Drives Price Evolution In the modern era of the oil market beginning roughly around 2000, there have been numerous demand shocks requiring a supply response from OPEC.  During the SARS outbreak in 2003, oil-market fundamentals at the time were complicated by the sudden loss of Venezuelan output in December 2002 to a general strike, which lasted three months and removed more than 2mm b/d from the market, and the US invasion of Iraq on March 2003. Both of these supply-side shocks hit markets just as demand was being hit by SARS. This makes it difficult to extract a pure price response on the demand side to the SARS episode. In the modern era of the oil market beginning roughly around 2000, there have been numerous demand shocks requiring a supply response from OPEC. These including the 9/11 terror attacks in the US in 2001; the SARS outbreak in late 2002-03; the Global Financial Crisis in 2007-08; and the euro debt crisis in 2011-12 (Chart 5).3 Chart 5Demand Shocks Abound In 21st Century Chart 6OPEC Lost Key Members' Output During SARS Outbreak   OPEC 2.0’s goal – similar to OPEC’s goal before it – is to avoid an unintended inventory accumulation. Importantly, these demand shocks were accompanied by supply shocks – Venezuela's general strike; the US invasion of Iraq continues to play havoc with global supply; the BP Macondo blowout in the Gulf of Mexico in 2010; the Arab Spring and the loss of Libyan output in 2011 – all of which complicated OPEC’s decision making (Chart 6). Much of OPEC’s adjustment then and now is made by the Kingdom of Saudi Arabia (KSA), which functions as the central bank of the global oil market increasing and decreasing production to balance markets (Chart 7). Chart 7KSA Primarily Balances Markets During Supply, Demand Shocks OPEC 2.0’s goal – similar to OPEC’s goal before it – is to avoid an unintended inventory accumulation, which would push prices lower and severely alter the forward curves for the principal crude oil pricing benchmarks, WTI and Brent (Chart 8). Chart 8OPEC 2.0’s Goal: Avoid Unintended Inventory Accumulation Navigating The 2019-nCoV Outbreak Oil prices – like all commodity prices – are a function of supply and demand, which clear the market instantaneously (here and now), and across time as buyers and sellers contract for forward delivery. The relentless focus on the demand-side consequences of the 2019-nCoV outbreak is not helpful in determining how oil prices will trade going forward. Oil prices – like all commodity prices – are a function of supply and demand, which clear the market instantaneously (here and now), and across time as buyers and sellers contract for forward delivery. The discussion above is meant to highlight this, by recalling OPEC’s production management during various demand shocks, not just the SARS outbreak in 2003. OPEC then, and OPEC 2.0 now, is not forced to produce oil and export regardless of the physical realities it confronts. It can adjust production and exports in response to direct demand indications from its refinery buyers and traders lifting its crude oil. Demand slowdowns, all else equal, typically will show up in falling crack-spread differentials between refined products and crude oil prices (Chart 9).4 Chart 9Crack Spreads Inform Crude Oil Production Decisions It still is too early to gauge the extent of the fall-off in demand arising from 2019-nCoV, but it will become apparent in cracks and in OPEC 2.0 producers’ responses to lower refiner demand. Falling crack spreads inform crude oil producers they need to throttle back on production – refiners are not able to profitably run all the crude being made available to them and crude and product are backing up in inventory.  It still is too early to gauge the extent of the fall-off in demand arising from 2019-nCoV, but it will become apparent in cracks and in OPEC 2.0 producers’ responses to lower refiner demand, which will determine how much production they need to cut in order to balance the market. This will be done against a backdrop of supply concerns that are not too dissimilar to those prevailing during the 2003 SARS crisis – e.g., instability in Iraq and Iran that could threaten production, and the loss of Venezuelan exports. Bottom Line: Markets still are in the process of assessing how damaging 2019-nCoV will be for industrial commodity demand – oil, bulks and base metals, in particular. As has been the case in all such demand shocks, OPEC’s supply response (and now OPEC 2.0’s) will determine how deeply and for how long prices are impacted.   Robert P. Ryan Chief Commodity & Energy Strategist rryan@bcaresearch.com Hugo Bélanger Senior Analyst Commodity & Energy Strategy HugoB@bcaresearch.com   Commodities Round-Up Energy: Overweight Brent prices fell 8% since last Monday amid the coronavirus outbreak in China. The number of confirmed cases is rapidly expanding, reaching more than 6,000 as of Wednesday which surpasses the trajectory of SARS in the first month of the outbreak in 2003. Nonetheless, the fatality rate remains below that of SARS, estimated at less than 3% vs. ~ 10% for SARS. Separately, the WCS discount to WTI averaged -$23/bbl this month. This is in line with our view that the discount would drop below -$20/bbl in 1Q20. This level is appropriate to incentivize additional rail transportation to the US. We expect the discount will remain close to current levels and for crude-by-rail volumes to pick up this year (Chart 10). Base Metals: Neutral Base metals have been severely impacted this week by the coronavirus outbreak – copper, aluminum, zinc, and lead are down 9%, 4%, 9%, and 5%. A prolonged slowdown in China’s economic activity – the driver of the global industrial activity recovery we expect – would plunge metals’ prices. China’s base metal consumption more than doubled since 2003. Thus, the potential impact of 2019-nCoV is much larger compared to SARS and market participants are pricing in the probability of damaging scenarios to global growth. This explains the pronounced decline in metals’ prices this year vs. 2003. Precious Metals: Neutral Gold was one of the few commodities in the green since last week. The yellow metal rose 1% since last Monday, supported by renewed safe-haven demand flows. Gold and the USD have been rising simultaneously amid the virus outbreak, which is typical of uncertain periods. The spectrum of possible outcomes is wide and negatively skewed. This warrants protection through safe-haven assets. We remain strategically long gold as a portfolio hedge. Our recommendation is up 28% since inception. Ags/Softs: Underweight Corn markets focused on USDA reports of rising exports, highlighted by the sale of 124,355 MT to Mexico. CBOT March corn futures were up 6% Tuesday, reversing earlier losses Monday. Beans remain under pressure, as traders await tangible evidence that China will go ahead with purchases announced in the so-called phase-one deal negotiated between the US and China (Chart 11).  Chart 10WCS Discount Under Pressure Chart 11Markets Waiting For China Demand   Footnotes 1     The US Centers for Disease Control and Prevention’s 2019-nCoV website highlights the marked differences between China’s response to the current coronavirus outbreak vs the 2003 SARS outbreak. One notable response by the Chinese government this time around – besides the rapid lockdown on travel – has been the alacrity with which officials posted the genome for the virus to a global research database, which allowed US researchers to quickly compare it to the strain they isolated. Separately, Reuters reported Australian researchers were able to grow the virus in a lab, which could accelerate development of a vaccine.  2     Distillates comprise the so-called middle of the refined barrel, and include jet fuel, diesel fuel and heating oil (also known as gasoil). These are primarily associated with industrial markets – mining and transportation, e.g. – and are key barometers of economic activity generally.  3     The "modern" era for oil began roughly in 2000, when oil prices became a random walk. WTI prices were mean-reverting from 1986 to roughly 2000, then became a random walk. Please see Helyette Geman, (2007), "Mean Reversion versus Random Walk in Oil and Natural Gas Prices," in Advances in Mathematical Finance, Birkhäuser, Boston; and Haidar, I. and C.R. Wolff, "Forecasting crude oil price (revisited)," The proceeding of the 30th USAEE Conference, Washington , D.C. USA. 9-12 October, 2011. 4     The “crack spread” is the USD/bbl difference between refined-product prices and crude-oil prices. It represents the gross margin of refiners.   Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades TRADE RECOMMENDATION PERFORMANCE IN 2019 Q4 Commodity Prices and Plays Reference Table Trades Closed in 2019 Summary of Closed Trades

Special Report

Highlights BCA’s “Golden Rule of Bond Investing” framework, which links developed economy government bond returns to central bank policy rate “surprises” versus market expectations, also works in China. The relationship between unexpected changes in China’s de facto short-term policy rate and government bond yields has been surprisingly strong over the past decade. Any additional easing by the PBoC this year is likely to be focused on reducing lending rates to the real economy, not interbank rates (which drive government bond yields). As such, yields at the short-end are likely to be flat until later this year at the earliest, whereas yields at the long-end are likely to move modestly higher, at most. The persistent historical gap between economic growth and bond yields in China makes it difficult to forecast the structural outlook for yields using conventional methods. To the extent that Chinese policymakers succeed at shifting the drivers of growth from investment to consumption, we believe that bond yields are more likely to structurally rise than fall. Over the coming 6-12 months, investors should underweight Chinese government bonds versus Chinese equities and onshore corporate bonds. Within a regional government bond portfolio, however, investors should overweight USD-hedged China versus US and developed markets ex-US, as well as in unhedged terms. Feature Last year’s inclusion of Chinese onshore government and policy bank bonds in the Bloomberg Barclays Global Aggregate Index was a significant milestone of China’s journey to internationalize its capital markets. Other bond benchmark providers have since followed suit, highlighting that the trend of increased passive exposure to Chinese assets is likely to continue. Over the past year, the bulk of the market discussion concerning the addition of China to the major bond indices has focused on estimating the size of potential capital inflows that could be triggered and the related impact on onshore bond yields. By contrast, comparatively little work has been done to analyze the core drivers of Chinese government bond yields, and how they compare to the factors that influence yields in the developed markets that dominate the bond indices. This Special Report attempts to fill a hole in the analysis of Chinese bonds. This Special Report attempts to fill that hole in the analysis of Chinese bonds. We look at the predictability of China’s government bond market through the lens of BCA’s “golden rule” framework, and find a surprisingly strong relationship between changes in China’s de facto short-term policy rate and government bond yields. We then present our cyclical (6-12 month) and secular outlooks for government yields given this relationship, and conclude by presenting four specific investment recommendations pertaining to China’s fixed-income market with two audiences in mind: mainland/onshore investors who are focused on returns in unhedged RMB terms, and global fixed-income investors who are primarily focused on hedged US-dollar regional bond exposure. The Golden Rule Of Bond Investing, With Chinese Characteristics In a July 2018 Special Report,1 BCA’s Chief US Bond Strategist, Ryan Swift, elegantly distilled the cyclical US government bond call into a simple question: During the next 12-months, will the Federal Reserve move interest rates by more or less than what is currently priced into the market? Chart 1The (US) Golden Rule Of Bond Investing In Practice Ryan argued that a predictive framework for US Treasury returns built around the answer to this question has historically worked so well that it should be referred to as the “Golden Rule of bond investing” (Chart 1). In a follow-up report, our Global Fixed Income Strategy service confirmed that the Golden Rule also largely works in non-US developed market economies, with the exception of Japan due to the absence of any meaningful fluctuation in policy rates over the past two decades.2 The Golden Rule provides a very strong framework to aid fixed-income investors with their cyclical (i.e. 6-12 month) asset allocation decisions, by quantitatively linking government bond returns relative to cash – in other words, the excess return earned by taking duration risk - to policy rate “surprises” compared to what is discounted in shorter-term money markets. The practical application is that a decision to allocate to longer-maturity government bonds is reduced to a bet on whether a central bank will adjust policy rates by more or less than the market expects. The first question we address in this report is to what degree does the Golden Rule apply in China (in yield space rather than in return space), along with an explanation of any differences that may exist. However, we must first note why the Golden Rule of bond investing works, particularly in the US. The first reason is that there is a strong relationship between the US 3-month T-bill rate and Treasury yields of all other maturities. Conceptually, all fixed income investors have a choice when buying US government bonds: they can purchase a 3-month Treasury bill and simply perpetually roll over the position as it matures, or they can purchase a Treasury bond of a longer maturity. This means that yields on longer maturity Treasury bonds simply reflect investor expectations for the average 3-month T-bill rate over the life of the bond, plus some positive risk premium to compensate for the inherent uncertainty of the path and tendency of short-term yields. This helps explain the close link between cyclical changes in 3-month T-bill rates and yields on longer maturity Treasurys. Chart 2In The US, The 3-Month T-Bill Rate Perfectly Tracks The Fed Funds Rate The second reason for the Golden Rule’s success is that there is a very tight relationship between the effective Fed funds rate and the 3-month T-bill rate. While it is the (higher) discount rate that is the theoretical no-arbitrage ceiling for the 3-month rate, in practice T-bill rates trade extremely close to the Fed funds rate (Chart 2). This means that Fed funds rate “surprises” (relative to traded market expectations) are akin to surprises in the 3-month rate, which in turn strongly influence the expected future path of short-term interest rates and thus yields on longer maturity Treasurys. In China, we noted in a February 2018 Special Report3 that the 7-day interbank repo rate is now the de jure short-term policy rate in China following the establishment of an interest rate corridor system in 2015. Chart 3 presents our first test of the Golden Rule in China (in yield space rather than in return space), by plotting the annual change in the level of Chinese government bond yields alongside the 7-day repo rate “surprise” over the past year from 2010 to the present. Here, we use the first principal component of zero coupon Chinese government bond yields to represent the average level of yields (rather than selecting a particular maturity), and we use the 12-month RMB swap rate (versus 7-day repo) to represent market expectations for the policy rate. The chart highlights that the fit is good, as measured by a 50% R-squared between the two series. However, deviations in the relationship do exist, with the most notable exception having occurred in 2017: Chinese government bond yields rose considerably more than what the annual surprise in the 7-day repo rate would have suggested. Chart 3In China, The Golden Rule Works Decently Well Using 7-Day Repo... Chart 4...And Extremely Well Using 3-Month SHIBOR Chart 4 helps resolve a good portion of the 2017 discrepancy, and clarifies the link between Chinese monetary policy and government bond yields. Chart 4 is similar to Chart 3, except that it replaces the 7-day repo rate surprise with that of 3-month SHIBOR (which trades very closely to the 3-month repo rate). The chart illustrates an even closer fit between the two series (with an R-squared close to 80%), and shows that the 3-month SHIBOR surprise does a meaningfully better job at explaining the 2017 rise in Chinese government bond yields. The Golden Rule of bond investing works surprisingly well in China. The fact that the annual surprise in 3-month SHIBOR has done a better job at predicting changes in bond yields over the past decade underscores that the 3-month repo rate is the de facto short-term policy rate in China, a point that we have made in several previous reports. We have noted that the spike in the 3-month/7-day repo rate spread that occurred in late-2016 and lasted until mid-2018 happened because of China’s crackdown on shadow banking activity. This crackdown caused a funding squeeze for China’s small & medium banks, which caused a material rise in lending rates and government bond yields. This episode highlights that future changes in the 3-month repo rate are likely to reflect both underlying changes in net liquidity provided to large commercial banks (measured by the 7-day repo rate), and any dislocations in the interbank market that have the potential to push up lending rates and government bond yields. Bottom Line: BCA’s “Golden Rule” framework, which links developed economy government bond returns to central bank policy rate “surprises” versus market expectations, works for China as well – using the correct measure of the PBOC policy rate. This provides a useful investment framework for Chinese government bonds, which are now significant part of major global bond market benchmarks. The Cyclical Outlook For Chinese Government Bond Yields Given the establishment of the relationship between Chinese short-term interbank rates and government bond yields detailed above, we are now able to more precisely discuss the likely cyclical trajectory of Chinese government bond yields as a function of Chinese monetary policy. Two opposing forces have the potential to affect China’s government bond market this year. The first, a stabilization and modest rebound in Chinese economic activity, may exert upward pressure on yields due to expectations of eventual policy tightening. The second, continued attempts by the PBoC to ease corporate lending rates, may exert downward pressure on yields as it will reflect not just easy but easier monetary conditions. Yields at the long-end are likely to move modestly higher this year, at most. For investors, the raises the obvious question of whether Chinese government bond yields are likely to move up, down, or trend sideways this year. In our view, yields at the short-end are likely to be flat until later this year at the earliest, whereas yields at the long-end are likely to move modestly higher, at most. Yields at the short-end of China’s government bond curve are likely to stay flat for most of this year. There are two reasons why yields at the short-end of China’s government bond curve are likely to stay flat for most of this year. The first is that the PBoC is generally a reactive central bank and has historically lagged a pickup in economic activity, as illustrated in Chart 5. The chart shows the historical path of 3-month SHIBOR in the year following a bottom in economic activity in 2009, 2012, and 2015, and makes it clear that there has been no precedent for a significant rise in interbank rates in the first nine months of an economic recovery. The 2012 episode did see a very sharp rise in 3-month SHIBOR once the PBoC shifted into tightening mode, but we doubt that this experience will be repeated again unless economic growth accelerates much more aggressively than we expect. The second reason why we expect yields at the short-end of the curve to remain muted this year is because any additional easing by the PBoC is likely to be focused on reducing corporate lending rates, not interbank rates. Chart 6 highlights that while there is a strong correlation between changes in Chinese government bond yields and average lending rates in the economy, the former leads the latter. In the past, this relationship has existed because changes in interbank rates have coincided with reductions in the now obsolete benchmark lending rate, with the former usually occurring earlier than the latter. But in a scenario where the PBoC reduces the loan prime rate (LPR) and keeps net banking sector liquidity roughly constant, the extremely tight relationship shown in Chart 4 suggests that short-term bond yields are unlikely to be affected by a reduction in lending rates. Any meaningful decline in short-term yields below short-term interbank rates would simply prompt banks to stop buying these bonds. Chart 5The PBoC Is Generally A Reactive Central Bank Chart 6Average Lending Rates Lag Short-Term Bond Yields Chart 7China's Yield Curve Is Generally Pro-Cyclical Additional easing by the PBoC does have the potential to impact the long-end of the government bond curve if investors view these actions as a sign that interbank rates will remain low for some time. This view is reinforced by the fact that China’s yield curve is not particularly flat, and thus has room to move lower. However, Chart 7 also shows that China’s yield curve, defined here as the second principal component of zero coupon Chinese government bond yields, is positively correlated with the relative performance of investable Chinese equities. This suggests that there is a procyclical element to the curve. We suspect that this procyclical element will dominate a potential decline in expectations for future short-term interest rates, but that yields at the long-end are likely to move modestly higher this year, at most. Bottom Line: Any additional easing by the PBoC this year is likely to be focused on reducing lending rates to the real economy, not interbank rates (which drive government bond yields). As such, yields at the short-end are likely to be flat until later this year at the earliest, whereas yields at the long-end are likely to move modestly higher, at most. The Secular Outlook For Chinese Government Bond Yields A common approach to forecasting the likely structural trend for nominal government bond yields is to estimate the trajectory of real long-term potential output growth and to add the monetary authority’s inflation target. This framework is based on the idea that interest rates are in equilibrium when the cost of borrowing is roughly equal to nominal income growth, a condition that results in no change in the burden to service existing debt. Chart 8China's Potential Growth Is Likely To Trend Lower... Based on this framework, we would expect Chinese government bond yields to trend down over time, or possibly flat if the PBoC were to tolerate higher inflation over the coming decade. Chart 8 illustrates the IMF’s forecast of falling real potential growth in China over the coming several years, which is consistent with a shift in the composition of growth from investment to consumption as well as China’s looming demographic crisis. But Chart 9highlights an obvious problem with applying this framework to forecast the secular trend in Chinese government bond yields: over the past decade, yields have persistently averaged below actual nominal GDP growth, both in China and in the developed world. In the latter case, it is an open question whether this will continue to be true in the future, but in China’s case it is clear that government bond yields have little connection (in magnitude) to the pace of GDP growth. This reflects the longstanding strategy of Chinese policymakers to promote investment via persistently low interest rates, as has occurred in other manufacturing and export-oriented Asian economies (Chart 10). Chart 9...But Bond Yields Are Well Below GDP Growth, Just Like In Developed Markets Chart 10In Industrial Asian Economies, Low Bond Yields Are A Policy Choice   The persistent historical gap between economic growth and bond yields in China makes it difficult to forecast the structural outlook for yields using conventional methods, and largely limits us to inference. To the extent that Chinese policymakers succeed at shifting the drivers of growth from investment to consumption, bond yields are more likely to rise than fall over time. This is because as long as interest rates remain well below the pace of income growth, the incentive to excessively borrow (and invest) is likely to persist. Chart 11China Needs Higher Interest Rates, But Only To A Point However, even in a scenario where Chinese government bond yields structurally trend higher, we expect the rise to be modest. Chart 11 highlights that China’s “private sector” debt service ratio is extremely elevated, underscoring that the country’s ability to tolerate significantly higher bond yields is not strong. In addition, since 2015, China’s debt service ratio has been mostly flat despite rising a rising debt-to-GDP ratio, which has been achieved through lower short-term interest rates. To the extent that policymakers fail to make meaningful progress in shifting China’s growth drivers away from investment over the coming few years, lower (potentially sharply lower) bond yields would appear to be all but inevitable to cope with what would become a permanently growing drag on economic activity from the servicing of debt. For now, we would characterize this scenario as a risk to our base case view, but it is a risk that we will be closely monitoring over the coming years. Bottom Line: The persistent gap between Chinese nominal GDP growth and government bond yields is likely contributing to the problem of excessive leveraging. To the extent that Chinese policymakers succeed at shifting the drivers of growth from investment to consumption, bond yields are more likely to structurally rise than fall. Investment Conclusions Our analysis above points to four recommendations for investors over the coming year: Overweight Chinese stocks versus Chinese government bonds in RMB and USD terms Overweight Chinese onshore corporate bonds versus duration-matched Chinese government bonds in RMB terms Overweight 7-10 year USD-hedged Chinese government bonds versus their US and developed market (DM) counterparts For offshore US dollar-based investors, long 7-10 year Chinese government bonds in unhedged terms Regarding the first two recommendations, our view that yields are likely to be flat at the short-end and modestly higher at the long-end suggests that investors can expect total returns on the order of 2-3% from Chinese government bonds this year. Barring a major and lasting economic slowdown from the 2019-nCoV outbreak, we expect Chinese domestic and investable equities to outperform government securities over the coming 6-12 months. Onshore corporate bonds have a similar outlook: onshore spreads are pricing in (massively) higher default losses than we believe is warranted, meaning that they will outperform duration-matched government equivalents without any changes in yield. Chart 12Within Global Fixed-Income, Hedged Chinese 10-Year Yields Are Relatively Attractive Chart 13Unhedged Yield Spreads Predict Hedged Relative Performance Versus The US For global fixed-income investors, Charts 12-14 present USD-hedged 10-year Chinese government yields versus the US and DM/DM ex-US, along with the historical relative return profile of USD-hedged Chinese bonds versus hedged and unhedged returns. In hedged space, Chinese 10-year government bond yields are modestly attractive: 2.2% versus 1.6% in the US and 1.8% in DM ex-US. China’s historically low yield beta to the overall level of global 10-year bond yields (Chart 15) suggests that Chinese yields should perform well in 2020 – a year where we expect global bond yields to drift higher as economic growth rebounds. Combined with relatively attractive valuation, this bodes well for the relative performance of Chinese debt versus DM equivalents. A low yield beta against a backdrop of drifting higher global yields implies that longer-maturity Chinese government bonds will outperform their DM equivalents. Chart 14Unhedged Yield Spreads Predict Hedged Relative Performance Versus DM Chart 15China's Yield Beta Has Been Rising, But Is Still Japan-Like   We would also recommend longer-maturity Chinese government bonds in unhedged terms versus a USD-hedged global government bond portfolio. Chart 16 highlights that the relative return of this trade is strongly (negatively) linked to USD-CNY, and we expect further (albeit more modest) gains in RMB over the cyclical horizon. Chart 16Modest Further RMB Upside Means Unhedged Chinese Bonds Will Outperform As a final point, investors should note that today’s report is part of a heightened focus on China’s fixed income market, in terms of both forecasting fixed income returns and analyzing the cyclical and structural implications of the increasing investability of China’s financial markets. More research on this topic is likely to come in 2020 and beyond: Stay Tuned!   Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com         Footnotes 1    Please see US Bond Strategy Special Report "The Golden Rule Of Bond Investing," dated July 24, 2018, available at usbs.bcaresearch.com 2   Please see Global Fixed Income Strategy Special Report "The Global Golden Rule Of Bond Investing," dated September 25, 2018, available at gfis.bcaresearch.com 3   Please see China Investment Strategy Special Report "Seven Questions About Chinese Monetary Policy," dated February 22, 2018, available at cis.bcaresearch.com

Special Report

Dear clients, Please note that in next week’s China Macro And Market Review, we will include a section explaining our view on the coronavirus outbreak and its economic as well as financial market implications. We maintain our overweight stance on both Chinese investable and A-share equities, over a tactical (0-3 months) and cyclical (6-12 months) time horizon. Please stay tuned. Jing Sima, China Strategist   Highlights BCA’s “Golden Rule of Bond Investing” framework, which links developed economy government bond returns to central bank policy rate “surprises” versus market expectations, also works in China. The relationship between unexpected changes in China’s de facto short-term policy rate and government bond yields has been surprisingly strong over the past decade. Any additional easing by the PBoC this year is likely to be focused on reducing lending rates to the real economy, not interbank rates (which drive government bond yields). As such, yields at the short-end are likely to be flat until later this year at the earliest, whereas yields at the long-end are likely to move modestly higher, at most. The persistent historical gap between economic growth and bond yields in China makes it difficult to forecast the structural outlook for yields using conventional methods. To the extent that Chinese policymakers succeed at shifting the drivers of growth from investment to consumption, we believe that bond yields are more likely to structurally rise than fall. Over the coming 6-12 months, investors should underweight Chinese government bonds versus Chinese equities and onshore corporate bonds. Within a regional government bond portfolio, however, investors should overweight USD-hedged China versus US and developed markets ex-US, as well as in unhedged terms. Feature Last year’s inclusion of Chinese onshore government and policy bank bonds in the Bloomberg Barclays Global Aggregate Index was a significant milestone of China’s journey to internationalize its capital markets. Other bond benchmark providers have since followed suit, highlighting that the trend of increased passive exposure to Chinese assets is likely to continue. Over the past year, the bulk of the market discussion concerning the addition of China to the major bond indices has focused on estimating the size of potential capital inflows that could be triggered and the related impact on onshore bond yields. By contrast, comparatively little work has been done to analyze the core drivers of Chinese government bond yields, and how they compare to the factors that influence yields in the developed markets that dominate the bond indices. This Special Report attempts to fill a hole in the analysis of Chinese bonds. This Special Report attempts to fill that hole in the analysis of Chinese bonds. We look at the predictability of China’s government bond market through the lens of BCA’s “golden rule” framework, and find a surprisingly strong relationship between changes in China’s de facto short-term policy rate and government bond yields. We then present our cyclical (6-12 month) and secular outlooks for government yields given this relationship, and conclude by presenting four specific investment recommendations pertaining to China’s fixed-income market with two audiences in mind: mainland/onshore investors who are focused on returns in unhedged RMB terms, and global fixed-income investors who are primarily focused on hedged US-dollar regional bond exposure. The Golden Rule Of Bond Investing, With Chinese Characteristics In a July 2018 Special Report,1 BCA’s Chief US Bond Strategist, Ryan Swift, elegantly distilled the cyclical US government bond call into a simple question: During the next 12-months, will the Federal Reserve move interest rates by more or less than what is currently priced into the market? Chart 1The (US) Golden Rule Of Bond Investing In Practice Ryan argued that a predictive framework for US Treasury returns built around the answer to this question has historically worked so well that it should be referred to as the “Golden Rule of bond investing” (Chart 1). In a follow-up report, our Global Fixed Income Strategy service confirmed that the Golden Rule also largely works in non-US developed market economies, with the exception of Japan due to the absence of any meaningful fluctuation in policy rates over the past two decades.2 The Golden Rule provides a very strong framework to aid fixed-income investors with their cyclical (i.e. 6-12 month) asset allocation decisions, by quantitatively linking government bond returns relative to cash – in other words, the excess return earned by taking duration risk - to policy rate “surprises” compared to what is discounted in shorter-term money markets. The practical application is that a decision to allocate to longer-maturity government bonds is reduced to a bet on whether a central bank will adjust policy rates by more or less than the market expects. The first question we address in this report is to what degree does the Golden Rule apply in China (in yield space rather than in return space), along with an explanation of any differences that may exist. However, we must first note why the Golden Rule of bond investing works, particularly in the US. The first reason is that there is a strong relationship between the US 3-month T-bill rate and Treasury yields of all other maturities. Conceptually, all fixed income investors have a choice when buying US government bonds: they can purchase a 3-month Treasury bill and simply perpetually roll over the position as it matures, or they can purchase a Treasury bond of a longer maturity. This means that yields on longer maturity Treasury bonds simply reflect investor expectations for the average 3-month T-bill rate over the life of the bond, plus some positive risk premium to compensate for the inherent uncertainty of the path and tendency of short-term yields. This helps explain the close link between cyclical changes in 3-month T-bill rates and yields on longer maturity Treasurys. Chart 2In The US, The 3-Month T-Bill Rate Perfectly Tracks The Fed Funds Rate The second reason for the Golden Rule’s success is that there is a very tight relationship between the effective Fed funds rate and the 3-month T-bill rate. While it is the (higher) discount rate that is the theoretical no-arbitrage ceiling for the 3-month rate, in practice T-bill rates trade extremely close to the Fed funds rate (Chart 2). This means that Fed funds rate “surprises” (relative to traded market expectations) are akin to surprises in the 3-month rate, which in turn strongly influence the expected future path of short-term interest rates and thus yields on longer maturity Treasurys. In China, we noted in a February 2018 Special Report3 that the 7-day interbank repo rate is now the de jure short-term policy rate in China following the establishment of an interest rate corridor system in 2015. Chart 3 presents our first test of the Golden Rule in China (in yield space rather than in return space), by plotting the annual change in the level of Chinese government bond yields alongside the 7-day repo rate “surprise” over the past year from 2010 to the present. Here, we use the first principal component of zero coupon Chinese government bond yields to represent the average level of yields (rather than selecting a particular maturity), and we use the 12-month RMB swap rate (versus 7-day repo) to represent market expectations for the policy rate. The chart highlights that the fit is good, as measured by a 50% R-squared between the two series. However, deviations in the relationship do exist, with the most notable exception having occurred in 2017: Chinese government bond yields rose considerably more than what the annual surprise in the 7-day repo rate would have suggested. Chart 3In China, The Golden Rule Works Decently Well Using 7-Day Repo... Chart 4...And Extremely Well Using 3-Month SHIBOR Chart 4 helps resolve a good portion of the 2017 discrepancy, and clarifies the link between Chinese monetary policy and government bond yields. Chart 4 is similar to Chart 3, except that it replaces the 7-day repo rate surprise with that of 3-month SHIBOR (which trades very closely to the 3-month repo rate). The chart illustrates an even closer fit between the two series (with an R-squared close to 80%), and shows that the 3-month SHIBOR surprise does a meaningfully better job at explaining the 2017 rise in Chinese government bond yields. The Golden Rule of bond investing works surprisingly well in China. The fact that the annual surprise in 3-month SHIBOR has done a better job at predicting changes in bond yields over the past decade underscores that the 3-month repo rate is the de facto short-term policy rate in China, a point that we have made in several previous reports. We have noted that the spike in the 3-month/7-day repo rate spread that occurred in late-2016 and lasted until mid-2018 happened because of China’s crackdown on shadow banking activity. This crackdown caused a funding squeeze for China’s small & medium banks, which caused a material rise in lending rates and government bond yields. This episode highlights that future changes in the 3-month repo rate are likely to reflect both underlying changes in net liquidity provided to large commercial banks (measured by the 7-day repo rate), and any dislocations in the interbank market that have the potential to push up lending rates and government bond yields. Bottom Line: BCA’s “Golden Rule” framework, which links developed economy government bond returns to central bank policy rate “surprises” versus market expectations, works for China as well – using the correct measure of the PBOC policy rate. This provides a useful investment framework for Chinese government bonds, which are now significant part of major global bond market benchmarks. The Cyclical Outlook For Chinese Government Bond Yields Given the establishment of the relationship between Chinese short-term interbank rates and government bond yields detailed above, we are now able to more precisely discuss the likely cyclical trajectory of Chinese government bond yields as a function of Chinese monetary policy. Two opposing forces have the potential to affect China’s government bond market this year. The first, a stabilization and modest rebound in Chinese economic activity, may exert upward pressure on yields due to expectations of eventual policy tightening. The second, continued attempts by the PBoC to ease corporate lending rates, may exert downward pressure on yields as it will reflect not just easy but easier monetary conditions. Yields at the long-end are likely to move modestly higher this year, at most. For investors, the raises the obvious question of whether Chinese government bond yields are likely to move up, down, or trend sideways this year. In our view, yields at the short-end are likely to be flat until later this year at the earliest, whereas yields at the long-end are likely to move modestly higher, at most. Yields at the short-end of China’s government bond curve are likely to stay flat for most of this year. There are two reasons why yields at the short-end of China’s government bond curve are likely to stay flat for most of this year. The first is that the PBoC is generally a reactive central bank and has historically lagged a pickup in economic activity, as illustrated in Chart 5. The chart shows the historical path of 3-month SHIBOR in the year following a bottom in economic activity in 2009, 2012, and 2015, and makes it clear that there has been no precedent for a significant rise in interbank rates in the first nine months of an economic recovery. The 2012 episode did see a very sharp rise in 3-month SHIBOR once the PBoC shifted into tightening mode, but we doubt that this experience will be repeated again unless economic growth accelerates much more aggressively than we expect. The second reason why we expect yields at the short-end of the curve to remain muted this year is because any additional easing by the PBoC is likely to be focused on reducing corporate lending rates, not interbank rates. Chart 6 highlights that while there is a strong correlation between changes in Chinese government bond yields and average lending rates in the economy, the former leads the latter. In the past, this relationship has existed because changes in interbank rates have coincided with reductions in the now obsolete benchmark lending rate, with the former usually occurring earlier than the latter. But in a scenario where the PBoC reduces the loan prime rate (LPR) and keeps net banking sector liquidity roughly constant, the extremely tight relationship shown in Chart 4 suggests that short-term bond yields are unlikely to be affected by a reduction in lending rates. Any meaningful decline in short-term yields below short-term interbank rates would simply prompt banks to stop buying these bonds. Chart 5The PBoC Is Generally A Reactive Central Bank Chart 6Average Lending Rates Lag Short-Term Bond Yields Chart 7China's Yield Curve Is Generally Pro-Cyclical Additional easing by the PBoC does have the potential to impact the long-end of the government bond curve if investors view these actions as a sign that interbank rates will remain low for some time. This view is reinforced by the fact that China’s yield curve is not particularly flat, and thus has room to move lower. However, Chart 7 also shows that China’s yield curve, defined here as the second principal component of zero coupon Chinese government bond yields, is positively correlated with the relative performance of investable Chinese equities. This suggests that there is a procyclical element to the curve. We suspect that this procyclical element will dominate a potential decline in expectations for future short-term interest rates, but that yields at the long-end are likely to move modestly higher this year, at most. Bottom Line: Any additional easing by the PBoC this year is likely to be focused on reducing lending rates to the real economy, not interbank rates (which drive government bond yields). As such, yields at the short-end are likely to be flat until later this year at the earliest, whereas yields at the long-end are likely to move modestly higher, at most. The Secular Outlook For Chinese Government Bond Yields A common approach to forecasting the likely structural trend for nominal government bond yields is to estimate the trajectory of real long-term potential output growth and to add the monetary authority’s inflation target. This framework is based on the idea that interest rates are in equilibrium when the cost of borrowing is roughly equal to nominal income growth, a condition that results in no change in the burden to service existing debt. Chart 8China's Potential Growth Is Likely To Trend Lower... Based on this framework, we would expect Chinese government bond yields to trend down over time, or possibly flat if the PBoC were to tolerate higher inflation over the coming decade. Chart 8 illustrates the IMF’s forecast of falling real potential growth in China over the coming several years, which is consistent with a shift in the composition of growth from investment to consumption as well as China’s looming demographic crisis. But Chart 9highlights an obvious problem with applying this framework to forecast the secular trend in Chinese government bond yields: over the past decade, yields have persistently averaged below actual nominal GDP growth, both in China and in the developed world. In the latter case, it is an open question whether this will continue to be true in the future, but in China’s case it is clear that government bond yields have little connection (in magnitude) to the pace of GDP growth. This reflects the longstanding strategy of Chinese policymakers to promote investment via persistently low interest rates, as has occurred in other manufacturing and export-oriented Asian economies (Chart 10). Chart 9...But Bond Yields Are Well Below GDP Growth, Just Like In Developed Markets Chart 10In Industrial Asian Economies, Low Bond Yields Are A Policy Choice   The persistent historical gap between economic growth and bond yields in China makes it difficult to forecast the structural outlook for yields using conventional methods, and largely limits us to inference. To the extent that Chinese policymakers succeed at shifting the drivers of growth from investment to consumption, bond yields are more likely to rise than fall over time. This is because as long as interest rates remain well below the pace of income growth, the incentive to excessively borrow (and invest) is likely to persist. Chart 11China Needs Higher Interest Rates, But Only To A Point However, even in a scenario where Chinese government bond yields structurally trend higher, we expect the rise to be modest. Chart 11 highlights that China’s “private sector” debt service ratio is extremely elevated, underscoring that the country’s ability to tolerate significantly higher bond yields is not strong. In addition, since 2015, China’s debt service ratio has been mostly flat despite rising a rising debt-to-GDP ratio, which has been achieved through lower short-term interest rates. To the extent that policymakers fail to make meaningful progress in shifting China’s growth drivers away from investment over the coming few years, lower (potentially sharply lower) bond yields would appear to be all but inevitable to cope with what would become a permanently growing drag on economic activity from the servicing of debt. For now, we would characterize this scenario as a risk to our base case view, but it is a risk that we will be closely monitoring over the coming years. Bottom Line: The persistent gap between Chinese nominal GDP growth and government bond yields is likely contributing to the problem of excessive leveraging. To the extent that Chinese policymakers succeed at shifting the drivers of growth from investment to consumption, bond yields are more likely to structurally rise than fall. Investment Conclusions Our analysis above points to four recommendations for investors over the coming year: Overweight Chinese stocks versus Chinese government bonds in RMB and USD terms Overweight Chinese onshore corporate bonds versus duration-matched Chinese government bonds in RMB terms Overweight 7-10 year USD-hedged Chinese government bonds versus their US and developed market (DM) counterparts For offshore US dollar-based investors, long 7-10 year Chinese government bonds in unhedged terms Regarding the first two recommendations, our view that yields are likely to be flat at the short-end and modestly higher at the long-end suggests that investors can expect total returns on the order of 2-3% from Chinese government bonds this year. Barring a major and lasting economic slowdown from the 2019-nCoV outbreak, we expect Chinese domestic and investable equities to outperform government securities over the coming 6-12 months. Onshore corporate bonds have a similar outlook: onshore spreads are pricing in (massively) higher default losses than we believe is warranted, meaning that they will outperform duration-matched government equivalents without any changes in yield. Chart 12Within Global Fixed-Income, Hedged Chinese 10-Year Yields Are Relatively Attractive Chart 13Unhedged Yield Spreads Predict Hedged Relative Performance Versus The US For global fixed-income investors, Charts 12-14 present USD-hedged 10-year Chinese government yields versus the US and DM/DM ex-US, along with the historical relative return profile of USD-hedged Chinese bonds versus hedged and unhedged returns. In hedged space, Chinese 10-year government bond yields are modestly attractive: 2.2% versus 1.6% in the US and 1.8% in DM ex-US. China’s historically low yield beta to the overall level of global 10-year bond yields (Chart 15) suggests that Chinese yields should perform well in 2020 – a year where we expect global bond yields to drift higher as economic growth rebounds. Combined with relatively attractive valuation, this bodes well for the relative performance of Chinese debt versus DM equivalents. A low yield beta against a backdrop of drifting higher global yields implies that longer-maturity Chinese government bonds will outperform their DM equivalents. Chart 14Unhedged Yield Spreads Predict Hedged Relative Performance Versus DM Chart 15China's Yield Beta Has Been Rising, But Is Still Japan-Like   We would also recommend longer-maturity Chinese government bonds in unhedged terms versus a USD-hedged global government bond portfolio. Chart 16 highlights that the relative return of this trade is strongly (negatively) linked to USD-CNY, and we expect further (albeit more modest) gains in RMB over the cyclical horizon. Chart 16Modest Further RMB Upside Means Unhedged Chinese Bonds Will Outperform As a final point, investors should note that today’s report is part of a heightened focus on China’s fixed income market, in terms of both forecasting fixed income returns and analyzing the cyclical and structural implications of the increasing investability of China’s financial markets. More research on this topic is likely to come in 2020 and beyond: Stay Tuned!   Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com         Footnotes 1    Please see US Bond Strategy Special Report "The Golden Rule Of Bond Investing," dated July 24, 2018, available at usbs.bcaresearch.com 2   Please see Global Fixed Income Strategy Special Report "The Global Golden Rule Of Bond Investing," dated September 25, 2018, available at gfis.bcaresearch.com 3   Please see China Investment Strategy Special Report "Seven Questions About Chinese Monetary Policy," dated February 22, 2018, available at cis.bcaresearch.com Cyclical Investment Stance Equity Sector Recommendations

Special Report

Highlights Historically, the stocks with the lowest volatility have outperformed the stocks with the highest volatility, a trend that defies some of the most fundamental theories in finance. The low-volatility factor has outperformed the market in absolute return terms, with a substantial reduction in downside risk, in every major equity market. Compensation structure, benchmarking, analyst bias, and the preference for lottery-like stocks are all plausible explanations for why the low-volatility anomaly persists. Shifting some exposure from bonds to minimum-volatility equities might be an attractive way to keep returns high while remaining hedged against downside risk in a world of low interest rates. Feature Chart 1The Low-Volatility Anomaly Conventional wisdom suggests that achieving the right mix between risk and return requires a tradeoff: Take on too little risk, and returns will be subpar, but take on too much, and a large loss becomes likely. However, the empirical evidence shows that no such tradeoff exists in the equity market: Historically, the return of stocks with the lowest volatility has been better than the return of stocks with the highest volatility (Chart 1, top panel). Even when using a measure of systematic risk such as beta, there appears to be very little relationship between risk and return – a result that emphatically contradicts what is predicted by the CAPM (Chart 1, bottom panel). The success of low risk stocks – a well-documented phenomenon known as the low-volatility anomaly1  – challenges some of the most fundamental premises of finance and economics. After all, how could taking on less risk result in better returns? In this report, we dive into this anomaly, with the intent of answering the following questions: What kind of portfolios can exploit this anomaly? What are the risk/return characteristics to this factor and what sectors is it exposed to? Why does this factor work? How can investors use the low-volatility factor in their asset allocation process? To answer these questions we explore the historical performance of the MSCI minimum-volatility index. Additionally, we explore the academic research surrounding the low-volatility anomaly. Finally, we look into how low-volatility equities have performed as a hedge for a global equity portfolio when compared to government bonds. Low Volatility Vs. Minimum Volatility There are two types of portfolios that are generally used to exploit the low-volatility anomaly: Low-volatility portfolios are built first by sorting the stock universe according to the stocks' trailing standard deviation, and then by buying the stocks with the lowest standard deviation. Usually the index buys the bottom quintile of stocks ranked by volatility. Minimum-volatility portfolios are built through an optimization procedure, by which funds are allocated to the stock mix that would have minimized the historical volatility of a portfolio (subject to certain constraints). Chart 2No Dramatic Difference Between Low Vol And Min Vol The main advantage of minimum-volatility portfolios over low-volatility portfolios is that they do not consider only low-volatility stocks but also stocks with low covariance between each other. However, the construction of these portfolios also requires estimating large covariance matrices, which are prone to a significant degree of noise, and thus often have to be adjusted by statistical methods.2 That being said, there is little performance difference between these two methodologies in practice, though minimum-volatility portfolios do tend to be much more constrained than low-volatility portfolios (Chart 2). In this report we will focus on the more popular MSCI minimum-volatility portfolios, given that their historical data is more readily available.   Risk-Return Characteristics Of Minimum Volatility At the global level, minimum volatility has outperformed not only the market since 1990, but also the most popular equity factors, with the exception of momentum (Chart 3, top panel). The outperformance relative to the benchmark has proved to be robust, as minimum volatility has beaten the returns for the benchmark in the biggest developed markets as well as emerging markets for almost two decades (Chart 3, bottom panel). The most attractive feature of minimum volatility is the significant reduction in risk it provides. Since 1988, the annualized volatility of minimum volatility has been 10%, a considerable improvement vis-à-vis the market and relative to other popular equity factors (Chart 4, top panel). Meanwhile, even though return skew of minimum volatility has been more negative than the benchmark, minimum volatility achieved a substantial reduction in tail risk with a 10% conditional VaR of only 5% (Chart 4, middle and bottom panel). Chart 3Min Vol Outperformance Is Broad-Based Chart 4Min Vol Provides A Substantial Reduction In Risk In addition to its risk reduction, minimum volatility also has a countercyclical relative return profile, outperforming during bear markets, and underperforming slightly during bull markets (Chart 5, top panel). This return profile occurs partly due to the sector skew of this factor, which overweights defensive sectors such as Utilities and Consumer Staples relative to the global benchmark3 (Chart 5, middle panel). At the global level, this defensive tilt exposes minimum volatility to significant duration risk, given that the stocks in this index tend to have bond-like properties (Chart 5, bottom panel). The negative relationship between interest rates and the low-volatility factor has been well documented by the academic literature: Some studies estimate that up to 80% of the outperformance of low-volatility equities can be explained by exposure to duration risk.4 Chart 5Global Min Vol Is Sensitive To Interest Rates Chart 6Min Vol Is Expensive     On average, minimum volatility also tends to overweight stocks with a higher dividend yield than the market (Chart 6, panel 1). This yield difference accounts for roughly a quarter of the return difference between minimum volatility and the benchmark since 1990. However, high dividend yield and minimum volatility are not synonymous: Over the past decade, minimum volatility has selected stocks that are more expensive than the benchmark, a stark difference from high dividend yield portfolios, which exclusively select very cheap stocks (Chart 6, panel 2). The current high overvaluation of minimum volatility relative to the market could spell bad news for this factor in the near future: While valuation and returns do not have a straightforward relationship, extreme levels of valuation relative to the benchmark have historically resulted in subsequent underperformance of this factor relative to the market (Chart 6, panels 3 and 4). Why Does The Low-Volatility Factor Work? Analyst Forecasts It has been shown empirically that equity analysts tend to have an optimistic bias, due to the need of the analyst to maintain good relationships with the companies they cover. Moreover, investors do not adjust for this bias, meaning that the stocks of companies that receive an overly optimistic forecast tend to rise initially when the forecast is released, but then mean-revert once the forecast does not materialize. Chart 7Analyst Bias Is Larger In High-Vol Stocks It seems that this dynamic might be more pervasive in high-volatility stocks. In their paper “When Sell-Side Analysts Meet High-Volatility Stocks: An Alternative Explanation for the Low-Volatility Puzzle,” Hsu et al. show that analysts’ positive bias is larger for more volatile stocks, even when adjusting for confounding variables like size, sector and country5 (Chart 7).  It is easy to see why this might be the case: An extremely optimistic bias on a low variance stock would look unrealistic to most investors. Thus, analysts are more prone to express a large positive bias on high variance stocks, where bias is harder to detect. Ultimately, this bias causes these stocks to become chronically overvalued. Salary Structure of Managers Chart 8Institutions Favor High Vol The payoff structure of fund managers has also been suggested as a possible cause of the low-volatility anomaly.6 The structure of compensation for fund managers often resembles an option: A bonus is granted if the portfolio returns are higher than a certain threshold. In such a structure, the compensation of portfolio managers resembles a call option: The probability of a payoff increases with volatility, creating an incentive to invest in high-volatility stocks.  However, this preference for high-volatility stocks will overbid them, causing them to underperform. There is some evidence that this is in fact the case. Once the effect of size is neutralized, stocks with high institutional ownership tend to be more volatile than stocks with low institutional ownership (Chart 8). Moreover, a study on the exposure of hedge funds to popular risk factors from 2000 to 2016, showed that hedge funds have a large short exposure to the anomaly, indicating that they favor high-volatility over low-volatility stocks.7   Lottery Stocks Chart 9The Low-Volatility Anomaly Is Most Prevalent In Lottery-Like Stocks Not all high-volatility stocks are chronic underperformers. In the paper “The Low Volatility Anomaly and the Preference for Gambling,” Hsu et al. identify that the low-volatility anomaly is largest in stocks with strong lottery-like characteristics (the authors define lottery-like as the stocks that had the highest maximum daily return the previous month)8 (Chart 9). Meanwhile, other high-volatility stocks are much more likely to have higher or equal returns than their low-volatility counterparts. Why is this the case? Investors tend to have a preference for “home-run” stocks, which have a large probability of a small loss, but a small probability of a large gain – a well-documented behavioral bias known as the long-shot bias.9 This bias might cause investors to overbid for lottery-like stocks, causing them to underperform as a cohort. This phenomenon might be related to the incentives in the money management industry. Flows to equity funds tend to be very skewed to the very best performing funds. This means that fund managers have a high incentive to invest in stocks that have the potential of an extremely high payoff. Benchmarking Chart 10Benchmarking Could Be To Blame For The Low-Vol Anomaly While the low-volatility anomaly has been found in other asset classes,10 it is interesting to note that the anomaly occurs only within asset classes and not across asset classes. Indeed, the traditional risk-return relationship is conserved when looking at a multi-asset universe (Chart 10, top panel). One possible solution to this puzzle might be the effect of benchmarking. Within an asset class, benchmarked managers are not only measured according to their return relative to their benchmark but also to their tracking risk (volatility of return difference). Under this structure, investors do not have a large incentive to exploit the alpha from low-volatility or low-beta stocks, since such a strategy would result in a relatively high tracking error.11 This high tracking error will make the excess return of low-volatility stocks relatively less attractive for benchmarked managers, even if these stocks are clearly superior in terms of raw risk-adjusted returns (Chart 10, bottom panel). How Can Minimum Volatility Be Used In Asset Allocation? Chart 11Sensitivity Of Min Vol To Interest Rates Has Increased In The Last Decade The interest rate sensitivity of low-volatility stocks has been a topic of significant interest for academic researchers. A study that attempted to measure this sensitivity found that equities in the lowest volatility decile have an interest-rate exposure equivalent to a 66% equity/34% bond portfolio.12 Moreover, this sensitivity has increased in recent years: Factor analysis shows that while the beta of the excess returns of minimum volatility to the equity market has remained constant, the beta to the bond market has increased significantly over the last decade13 (Chart 11, top panel). Interestingly this process seems to have accelerated as bond yields fell below dividend yields – a result which might arise because the market starts perceiving bonds, and low-volatility equities as close substitutes when they provide a similar cash flow (Chart 11, bottom panel). At first glance, this relatively high duration risk appears to be a red flag, particularly if you share our view that interest rates have reached a multi-year bottom. After all, an environment where interest rates rise would imply that minimum volatility would underperform global equities on a structural basis.  However, the sensitivity of minimum volatility to interest rates can be used to the advantage of an asset allocator. Specifically, in a world of low bond yields, it could be attractive for an investor to shift some of the exposure he or she has from bonds to minimum-volatility equities. Why would an investor do this? As we discussed in our October 2019 report, low bond yields will cause bonds to generate returns that are much lower than their historical averages. This means that using government bonds as a hedge for an equity portfolio is likely to result in a severe drag on performance.14 On the other hand, while minimum-volatility equities do not have the same hedging potency as bonds, their returns are much higher, which suggests that at the right allocation they could prove to be a better hedge. In a world of low bond yields, it could be attractive for an investor to shift some of the exposure he or she has from bonds to minimum-volatility equities. How has such a strategy worked historically? Chart 12 shows how allocating incremental amounts of global minimum-volatility equities to an equity portfolio compares to adding incremental amounts of global government bonds. Overall, allocating an additional 3% of minimum-volatility equities to a portfolio had the same effect as adding 1% of government bonds in terms of downside protection and volatility reduction, but with a much higher return. This effect was robust throughout the sample period. Consider a portfolio with 30% government bonds, 30% minimum volatility and 40% global equities. With the exception of the 1990-1994 period, this portfolio had roughly the same 10% conditional VaR in all sub periods as a portfolio with 40% government bonds and 60% equities, but a significantly higher return (Chart 13). Moreover, the volatility of the portfolio that includes minimum volatility has also been lower than the 60/40 portfolio since 2000. Chart 12Min Vol Provides Protection With A Relatively High Return Chart 13Min-Vol Protection Has Been Robust Bottom Line The low-volatility anomaly contradicts what is perhaps the most fundamental pillar in finance: The tradeoff between risk and return. As such, this anomaly might be the most puzzling inefficiency in markets. But will it persist? The evidence suggests that the anomaly is caused by strong institutional incentives, which are likely to continue. Thus, investors should strongly consider investing in low-volatility stocks in the future. However, the following points should be taken into account: Investors who are evaluated according to their information ratio should avoid low/minimum-volatility stocks, given that their large volatility difference with the benchmark will result in a relatively high tracking error. Relatively high valuations and an uptrend in interest rates may hurt the performance of low/minimum-volatility stocks relative to the broad equity market in the near future. However, the interest-rate exposure of low/minimum-volatility stocks might prove attractive to multi-asset investors. Specifically, investing in minimum-volatility equities and reducing bond exposure might be a way to boost returns while remaining hedged.   Juan Correa Ossa, CFA Senior Analyst juanc@bcaresearch.com   Footnotes 1  The discovery of the low-volatility anomaly predates the discovery of both the size and value anomalies. For more details, please see Michael C. Jensen, Fischer Black, and Myron S. Scholes, “The Capital Asset Pricing Model: Some Empirical Tests,” Studies in the Theory Of Capital Markets, Praeger Publishers Inc., 1972. 2 For more details on the construction of minimum-volatility portfolios, please see Tzee-man Chow, Jason C. Hsu,Li-Lan Kuo, and Feifei Li, “A Study of Low Volatility Portfolio Construction Methods,” The Journal of Portfolio Management, Vol. 40, No. 4, 2014. 3 However, research has shown that the low-volatility anomaly still holds within sectors, suggesting that its outperformance is not only a result of sector bias. For more details, please see Raul Leote de Carvalho, Majdouline Zakaria, Lu Xiao, and Pierre Moulin, “Low Risk Anomaly Everywhere - Evidence from Equity Sectors,” (November 19, 2014). 4 Please see Joost Driessen, Ivo Kuiper, Korhan Nazliben, and Robbert Beilo, “Does Interest Rate Exposure Explain the Low-Volatility Anomaly?” Journal of Banking and Finance, Vol. 103, 2019. 5 Please see Jason C. Hsu, Hideaki Kudoh and Toru Yamada, “When Sell-Side Analysts Meet High-Volatility Stocks: An Alternative Explanation for the Low-Volatility Puzzle,”  Journal Of Investment Management (JOIM), Second Quarter 2013. 6 Please see Nardin L. Baker and Robert A. Haugen, “Low Risk Stocks Outperform within All Observable Markets of the World,” (April 27, 2012). 7 Please see David Blitz, “Are Hedge Funds on the Other Side of the Low-Volatility Trade?”  The Journal of Alternatives Investments, Vol. 21, No. 1, Summer 2018. 8 Please see Jason C. Hsu and Vivek Viswanathan, “The Low Volatility Anomaly and the Preference for Gambling,” Risk-Based and Factor Investing, pages 291-303; (2015). 9 There is some debate as to whether the long-shot bias is truly irrational in financial markets, where payoffs and probabilities are not known with precision. 10 Researchers have found that the low-volatility anomaly exists in the government and corporate bond market. For more details please see, Raul Leote de Carvalho, Patrick Dugnolle, Lu Xiao, and Pierre Moulin, “Low-Risk Anomalies in Global Fixed Income: Evidence from Major Broad Markets,” The Journal of Fixed Income, vol. 23, No. 4, Spring 2014. 11 Please see Malcolm Baker, and Brendan Bradley, and Jeffrey Wurgler, “Benchmarks as Limits to Arbitrage: Understanding the Low-Volatility Anomaly,” Financial Analysts Journal, Vol. 67, No. 1, 2011. 12 Please see Joost Driessen, Ivo Kuiper, and Korhan Nazliben, and Robbert Beilo, “Does Interest Rate Exposure Explain the Low-Volatility Anomaly?” Journal of Banking and Finance, Vol. 103, 2019. 13 We calculate sensitivity for US stocks instead of global stocks due to the effects of currency returns. 14 Please see Global Asset Allocation Strategy Report “Safe Haven Review: A Guide To Portfolio Protection In The 2020s,” dated October 29 2019, available at gaa.bcaresearch.com.

Highlights The Wuhan coronavirus outbreak in China is now being priced into commodity markets, with comparisons to the 2003 SARS outbreak serving as an early benchmark.1 If it follows the SARS trajectory its impact likely will be limited, although oil demand could fall at the margin as global travel falls. The IMF expects growth in EM economies, the engine for commodity demand, to come in at 4.4% and 4.6% this year and next, respectively, down two-tenths of a percent from its previous forecast, but still up from 2019’s 3.7% rate. The Fund’s risk assessment tilts slightly to the upside, nonetheless, in the wake of global monetary and fiscal stimulus. We introduce our 2021 oil balances and price forecasts this week. We expect Brent crude oil to average $70/bbl next year, and for WTI to average $4/bbl below that. We are maintaining our $67/bbl Brent and $63/bbl WTI 2020 forecasts (Chart of the Week). Chart of the WeekCrude Oil Price Forecasts For 2020, 2021 Feature In its latest World Economic Outlook – Tentative Stabilization, Sluggish Recovery? – the IMF flags key risks to EM growth, which will continue to feed the economic policy uncertainty that dogs commodity demand.2 The Fund’s “downward revision primarily reflects negative surprises to economic activity in a few emerging market economies, notably India, which led to a reassessment of growth prospects over the next two years. In a few cases, this reassessment also reflects the impact of increased social unrest.” That said, the Fund sees the balance of risk slightly tilted to the upside versus its earlier assessment in October, in the wake of global monetary and fiscal stimulus. This is in line with our view that the effects of monetary stimulus – deployed over the better part of last year and still expected to remain accommodative this year – will boost growth this year. Our view remains tempered by risks we’ve been highlighting that keep political and economic policy uncertainty elevated – e.g., trade tensions, civil unrest, and the still-underappreciated risks to oil markets arising from US-Iran tensions and social unrest in Iraq, which remains high (Chart 2). The loss of 800k b/d from Libya is significant, but the world does not lack spare light-sweet crude oil production capacity – the US shales, in particular, abound in this type of crude oil. Chart 2Policy Uncertainty Will Trend Lower, But Continues To Dog Commodities Oil Fundamentals Improving As is typically the case, we expect global oil-demand growth this year will be led by EM economies. Crude oil fundamentals continue to favor higher prices: Production management and capital discipline will constrain the rate of growth of oil supplies, and, as discussed above, demand will benefit from policy stimulus globally (Chart 3). Oil demand growth will recover this year, following a lower-than-normal rate of just 830k b/d last year, based on the US EIA’s most recent estimates of historical consumption. We continue to expect demand to grow 1.4mm b/d this year.  For 2021, we expect growth of just under 1.5mm b/d, reaching 103.65mm b/d globally. For its part, the EIA’s estimating growth of 1.34mm and 1.37mm b/d for 2020 and 2021, respectively. As is typically the case, we expect global oil-demand growth this year will be led by EM economies, proxied by non-OECD oil consumption, of 1.26mm b/d. For next year, we expect EM demand growth to come in at 1.34mm b/d, or just over 90% of global oil consumption growth in 2021. On the supply side, we continue to expect OPEC 2.0 output to increase slightly in 2Q20 and return to levels consistent with its previous agreement to cut 1.2mm b/d of production. Our modeling also assumes this level of production remains flat for the rest of 2020. Chart 3Fundamental Supply-Demand Balances Support Higher Crude Oil Prices Next year, we assume the producer coalition led by the Kingdom of Saudi Arabia (KSA) and Russia to increase production by 350k b/d in 1H21. In addition, we gradually remove 300k b/d of KSA’s overcompliance of 400k b/d next year, which moves its crude oil output in 2021 to 9.94mm b/d vs 9.76mm b/d this year. For Russia, we anticipate an increase in its condensate production, which it lobbied for last year. This will put our estimate of Russia’s crude and condensate production at 11.4mm b/d in 2020 and 11.64mm b/d in 2021.3 Most of the production cuts realized by OPEC 2.0 – ~ 2mm b/d – come at the expense of Venezuela and Iran, both of which are under sanctions limiting their production imposed by the US. We are holding Venezuela’s production at ~ 700k b/d in 2021, and will be monitoring this closely for any indication it is significantly changing. For Iran, we are keeping its production at 2.10mm b/d this year and next, assuming US sanctions remain in place. Oil production in both countries could be impacted by the outcome of US elections in November, and right now this is a near-impossible call to make. US Shales: No Longer A Growth Story? We continue to see slower production growth in the US than the EIA, particularly in the shales, as we expect capital markets to continue to discipline shale producers by only funding those firms that are able to return capital to shareholders or to deliver steady and increasing dividends. In our modeling, total US onshore production this year and next is expected to rise 800k b/d, and 310k b/d for 2021. We also continue to expect drilled-but-uncompleted (DUC) wells to continue to make significant contributions to overall shale-oil production in the US. Indeed, we expect DUCs to continue to offset part of the decline implied by lower rig counts, as they require less capex than drilling and completing new wells. We add ~ 500k b/d of production from DUCs completion over 2020 and 2021. Future production will depend heavily on the Majors and on productivity and lateral length. Our US crude and condensate production estimates for 2020 and 2021 reflect these constraints, and the slowing rate of growth being imposed by capital markets. For 2020, we expect total US crude and condensate production of 13.16mm b/d, of which 9.20mm b/d will come from the main shale basins led by the Permian.4 Tighter Fundamentals, Steeper Backwardations Our fundamental supply-demand balances are tighter than those assumed by the US EIA and the Paris-based IEA (Table 1). We expect US crude and liquids production to grow 1.6mm b/d this year, and only 500k b/d next year. We see global production growing 1.15mm b/d and 1.39mm b/d in 2020 and 2021, respectively. With demand growing 1.4mm b/d and close to 1.5mm b/d in 2020 and 2021, respectively, against this supply backdrop, our balances point to a deficit this year vs. the surplus expected by the IEA  (Table 2 and Chart 4). Table 1Fundamentals Comparison Table 2BCA Global Oil Supply - Demand Balances (MMb/d, Base Case Balances) Chart 4BCA Research's Balances Estimates Point To Falling Inventories Chart 5Tighter Storage, Steeper Backwardation For this reason, we continue to anticipate a steepening in the Brent and WTI forward curves – i.e., more backwardation – which will support our long 2H20 Brent vs. short 2H21 Brent curve trade (Chart 5). As a result of the steeper backwardation, we expect higher volatility, and will be getting long 4Q20 Brent $65/bbl calls vs. short 4Q20 Brent $70/bbl calls (Chart 6). Bottom Line: We continue to expect crude oil markets to tighten, given persistent production restraint by OPEC 2.0, capital-market-imposed restraint on US shale-oil producers, and revived global demand growth in 2020 and 2021. The IMF’s assessment re the balance of risk being tilted to the upside, in the wake of global monetary stimulus, is broadly consistent with our maintained view. While we expect global policy uncertainty to fall following the so-called phase-one US-China trade deal and a definitive Brexit vote in the UK, geopolitical tension remains high, particularly in the Persian Gulf. Chart 6Steeper Backwardation To Higher Implied Volatility We will be getting long 4Q20 Brent $65/bbl calls vs. short 4Q20 Brent $70/bbl calls, in anticipation of higher volatility in the wake of lower inventories. As a result, we are keeping our 2020 Brent forecast at $67/bbl, and are expecting 2021 Brent to trade at $70/bbl; WTI is expected to trade $4/bbl below Brent this year and next, on average. At tonight’s close, we will be getting long 4Q20 Brent $65/bbl calls vs. short 4Q20 Brent $70/bbl calls, in anticipation of higher volatility in the wake of lower inventories.   Robert P. Ryan Chief Commodity & Energy Strategist rryan@bcaresearch.com Hugo Bélanger Senior Analyst Commodity & Energy Strategy HugoB@bcaresearch.com   Commodities Round-Up Energy: Overweight Brent prices traded sideways ~ $64/bbl since last Tuesday, dismissing the US and China phase-one agreement and disruptions to Libyan production and exports which could total as much as 800k b/d.  Over the weekend, concerns re the Wuhan coronavirus outbreak in China started being priced into commodities, particularly oil.  Separately, the US Treasury Department renewed Chevron’s waiver to operate in Venezuela for another three months.  The company is scheduled to export 1mm barrels of oil produced by PDVSA via a joint-venture, partially dodging US sanctions on Venezuelan oil.5  We expect the country’s output to stabilize close to its current level of 710 kb/d this year. Base Metals: Neutral On Tuesday Beijing reported more than 400 people had been infected with the Wuhan coronavirus, confirming person-to-person transmission of the virus. Concerns that a wider spread over the lunar New Year holidays starting this weekend will impact economic growth in the world’s top metal consumer brought copper prices down 1.8% on Tuesday.  Zinc reached two-month highs this week amidst concerns of low LME warehouses stocks, now close to their 20-year lows at 50,900 MT (Chart 7).  Supply concerns stemming from low iron ore stocked in China’s ports, along with good Chinese macro data, lifted iron-ore prices. Precious Metals: Neutral The US dollar is a key missing piece needed to propel gold prices higher from current levels. The 2.4% decline in the trade-weighted dollar index supported gold’s 5% increase since October 1, 2019 (Chart 8).  We expect the dollar to continue depreciating in 2020, as global growth rebounds and the Fed remains accommodative, keeping gold prices well bid.  Most precious metals have followed gold’s lead this year; palladium and platinum are up 17.63% and 3.15%, respectively. Chart 7 Chart 8 Ags/Softs:  Underweight CBOT Corn and soybeans futures traded lower on Tuesday as markets awaited evidence of China purchasing additional U.S. agricultural goods, fulfilling its commitment to buy $32 billion of agricultural goods over two years per the phase-one deal negotiated between China and the US earlier this month.  Corn traded lower, as US grain elevators have yet to confirm any Chinese buying.  Soybeans, further weakened by expectations of a massive harvest in rival exporter Brazil.  Wheat was the only ag posting gains early in the week on the back of strong Black Sea export demand.     Footnotes 1     Please see CDC SARS Response Timeline, published by the US Centers for Disease Control and Prevention.  The SARS outbreak was identified in February 2003 and lasted six months.  The CDC noted: “Globally, WHO received reports of SARS from 29 countries and regions; 8,096 persons with probable SARS resulting in 774 deaths. In the United States, eight SARS infections were documented by laboratory testing and an additional 19 probable SARS infections were reported.”  According to Chinese officials, there were 440 confirmed cases of the new coronavirus as of Wednesday; nine people were reported to have died thus far.  The World Health Organization met Wednesday to assess the Wuhan coronavirus outbreak.  The 2003 coronavirus outbreak was minor compared to the typical influenza outbreak: by way of comparison, every year there are an estimated one billion cases of influenza, resulting in 290,000 to 650,000 deaths, according to the International Federation of Pharmaceutical Manufacturers & Associations in Switzerland. 2               Economic policy uncertainty is a recurrent theme in our research.  It has been driving safe-haven demand for the USD and gold for months, as we recently discussed in Iran Responds To US Strike; Oil Markets Remain Taut.  It is available at ces.bcaresearch.com. 3     We use World Bank growth estimates to drive our EM demand forecasts.  Earlier this month, the Bank forecast EM GDP growth of 4.1% for 2020 and 4.3% for next year.  This will outpace last year’s growth rate of 3.5%. 4     US production growth, particularly in the Permian and Bakken basins, could be constrained by environmental restrictions, if state regulators crack down on the massive flaring occurring in both states.  Please see Lingering Oil-Demand Weakness Will Fade, published November 21, 2019, where we discuss this risk in more depth. 5     Please see Exclusive: PDVSA's partners act as traders of Venezuelan oil amid sanctions - documents, published by reuters.com January 13, 2020.   Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades TRADE RECOMMENDATION PERFORMANCE IN 2019 Q4 Commodity Prices and Plays Reference Table Trades Closed in 2019 Summary of Closed Trades