Commodities & Energy Sector

Special Report

Highlights Crude oil prices fell ~ 10% Friday after Russia refused to support additional production cuts agreed by OPEC in Vienna (Chart 1). As we go to press, Brent is trading close to $45/bbl and WTI is trading ~ $41/bbl. OPEC producers could implement the go-to strategy they’ve employed in the wake of past demand shocks and cut production on their own, in order to balance the market. That said, there are indications the Saudis will not shoulder the market-balancing role alone. Russian producers have consistently demanded relief from production restraints since 2017, when OPEC 2.0 took over balancing the market. With shale-oil producers on the back foot owing to parsimonious capital markets, Russia could finally be able to deliver the coup de grâce it has been waiting for. This supply shock hits the market as COVID-19 threatens demand globally. Whatever Russia’s intent – be it removing the near-certainty of a production cut, which it always agreed to in the past, or crippling US shale production – two-way risk has returned to these Vienna meetings. Feature Oil markets once again are faced with a possible price collapse – not unlike the swan dive seen when OPEC’s market-share war took Brent from more than $110/bbl in mid-2014 to $26/bbl by early 2016. The proximate aim of that market-share war – led by the Kingdom of Saudi Arabia (KSA) – was to significantly reduce the revenue Iran would receive when it returned to export markets, following its agreement with the US to end its nuclear program in 2015. Tanking oil prices was the most expedient way of accomplishing this. Secondarily, shale-oil producers also may have been targeted, although such a goal was never clearly articulated by KSA’s leadership. Chart 1Russia's Supply Shock Craters Brent, WTI Prices OPEC’s market-share war did thin the US oil-shale herd, but it did not destroy the industry. If anything, it forced shale-oil producers to focus on their best drilling prospects with their best rigs and crews. This produced a leaner more productive technology-driven cohort of drillers, which posted record production levels on a regular basis. Indeed, by the end of 2019, US production topped 12.9mm b/d – 8.2mm b/d of which was accounted for by shale-oil output – making the US the largest oil and gas producer in the world. The market-share war also brought KSA and Russia together in November 2016 as the putative leaders of OPEC 2.0. The sole mission of this unlikely coalition was to clear the global inventory overhang left in the wake of the market-share war by managing OPEC and non-OPEC production. Russia’s Coup de Grâce Managing global production and inventories with KSA – while US shale-oil producers continued to raise their output to new records regularly – never sat well with Russia’s oil producers.   Managing global production and inventories with KSA – while US shale-oil producers continued to raise their output to new records regularly – never sat well with Russia’s oil producers. Ahead of OPEC 2.0 meetings in Vienna, Russian oligarchs could be counted on to demand higher output levels, and President Vladimir Putin could be counted on to deliver something close to agreed production cuts in time to assuage markets. This semi-annual ritual came to resemble a tightly choreographed set-piece, which may have inured market participants to the oligarchs’ resolve to ultimately increase production levels. Russia certainly was well-prepared when it delivered Friday’s supply shock. Time will tell, but Friday’s breakdown in Vienna could be the coup de grâce Russia’s oligarchs have been waiting to deliver to US shale producers since the formation of OPEC 2.0. Or it could be a well-timed reminder that nothing in oil markets is certain – particularly Russian compliance with production-restraint agreements. The once-certain 11th-hour agreement to adhere to whatever production-cutting agreements OPEC 2.0 came up with is now gone. And with it, the high-probability bet that, regardless of the tensions leading up to the Vienna meetings, a production-management agreement would be delivered, and shale-oil producers would live to fight another day. Chart 2Russia, KSA Foreign Exchange Reserves Whatever the case, Russia certainly was well-prepared when it delivered Friday’s supply shock. It has steadily built its foreign-exchange reserves since the price collapse begun in 2014, which now stand at $446 billion, up 45% from their nadir of 2015 (Chart 2). KSA’s foreign-exchange reserves, on the other hand, fell sharply in the wake of the 2014 – 2016 market-share war and have languished at lower levels since. Chart 3Russia, KSA Per-Capita Income Still, the Kingdom is not without stout resources. It’s gross national income per capita is ~ 2x that of Russia’s (Chart 3), and its days-forward import cover expressed in terms of days of foreign reserves is similarly stout (Chart 4). Chart 4Russia, KSA Import Cover The economies of both KSA and Russia are exquisitely linked to Brent oil prices (Chart 5). So tempting another market-share or price war is a strategy that could not be sustained by either country for an extended period of time. Chart 5Russia, KSA GDP vs Brent Prices Chart 6Russia, KSA GDP Highly Sensitive To Brent Prices The End Of OPEC 2.0? Post-GFC, we estimate Russia’s real GDP elasticity to changes in oil prices is close to twice that of Saudi Arabia. This suggests Russia’s strategy could have dismal consequences for its economy. Oil markets will gnaw on Friday’s breakdown in Vienna, sorting out the signals that were missed in Russian messaging, and figuring out what happens next. Neither Russia nor KSA have the resources to wage an indefinite war of attrition with US shale producers. Both are highly dependent on oil revenues to sustain their economies (Chart 6). Of the two, Russia’s economy is more sensitive to Brent oil prices than KSA’s, as it markets more of its output in trading markets. Post-GFC, we estimate Russia’s real GDP elasticity to changes in oil prices is close to twice that of Saudi Arabia. This suggests Russia’s strategy could have dismal consequences for its economy. Russia’s $50/bbl fiscal breakeven price vs. KSA’s $84/bbl price might give Russia more staying power in the short run, but with per-capita income at roughly half that of Saudi citizens, it will not want to revisit the dire days of 2014-16 when its economy last suffered through an oil-price collapse.   Robert P. Ryan Chief Commodity & Energy Strategist rryan@bcaresearch.com  

Highlights OPEC 2.0 ministers continue to negotiate oil production cuts to replace those expiring this month. We expect cuts of 1mm b/d – perhaps more – extending to end-June, undertaken to offset COVID-19-induced demand destruction. Making the not-unreasonable assumptions of no change in US sanctions-related output losses – 1mm b/d in Venezuela and 2mm b/d in Iran – and that 1mm b/d of Libyan output stays offline, the 1mm b/d cut coming out of this week’s meeting in Vienna will push average 1H20 OPEC 2.0 outages – planned and unplanned – to ~ 5mm b/d. The US economy is growing ~ 2.7% p.a., suggesting the Fed’s surprise 50bp rate cut this week is aimed at reducing global economic policy uncertainty (GEPU), lowering its accompanying USD safe-haven demand, and guarding against a collapse in US money velocity (Chart of the Week). This will weaken the USD, thereby supporting EM incomes and oil demand. We continue to expect policymakers in China to overshoot on fiscal and monetary stimulus, as they scramble to deliver 6% pa growth this year. Feature In the wake of ongoing negotiations – right into today’s meeting in Vienna – we expect OPEC 2.0 to deliver a production cut of at least 1mm b/d for 2Q20. Maybe more. Gulf Cooperation Council (GCC) states have been lobbying for a large cut – 1mm b/d at least. The Kingdom of Saudi Arabia’s (KSA) consistently lobbied for such cuts, and was instrumental in achieving the 1.7mm b/d output reduction for this quarter when the coalition met at the end of last year in Vienna. KSA’s partner in OPEC 2.0, Russia, has been slow to support production cuts going into this week’s meeting, which is the stance it typically takes during these negotiations. Nonetheless, it did agree in December to cuts, and we expect they will do so again this week. After this go-round, we’re likely to see an agreement to meet in June to determine whether cuts should be extended and/or expanded.1 Chart of the WeekFed Rate Cut Meant To Reduce Uncertainty The 1mm b/d in planned outages for 2Q20 coming out of this week’s meetings would add to the ~ 4mm b/d of unplanned outages in Venezuela, Iran and Libya this year. If the producer coalition fails to agree to a significant output cut this week, we would expect a sell-off in crude oil that takes Brent prices below $50/bbl, and WTI into the mid-$40s (Chart 2). An agreement to remove at least 1mm b/d of output likely will push Brent into the mid-$50s and WTI into the low-$50s during in 2Q20. Assuming the COVID-19 outbreak subsides by then, we expect Brent to rally in 2H20, with prices trading above $60/bbl and WTI trading $4/bbl below that on average. We will be updating our supply-demand balances and forecasts when we get fresh historical data from the key agencies (EIA, OPEC and IEA). The 1mm b/d in planned outages for 2Q20 coming out of this week’s meetings would add to the ~ 4mm b/d of unplanned outages in Venezuela, Iran and Libya. If these persist to end-June, planned and unplanned OPEC 2.0 production outages would average more than 5mm b/d in 1H20 (Chart 3). Chart 2A Failure To Cut Production Would Push Benchmark Crudes Lower Chart 3Core OPEC 2.0 Will Agree Cuts On the demand side, the big global hit to growth from China in 1Q20 should be out of the system by the end of 1H20, assuming the COVID-19 outbreak does not shut down global commerce the way it did in China. We think the odds of such a shutdown are low, given such policies only can be implemented by a central government in which all power is consolidated in a ruling party. Besides, given the massive hit to China’s manufacturing – auto production was down 80% y/y in February, e.g. – such policies are unlikely to be recommended in all but the most dire of circumstances. We continue to expect Chinese policymakers to overshoot on their fiscal and monetary stimulus, as they scramble to get 2020 GDP growth back above a 6% p.a. rate. Our view aligns with BCA’s China Investment Strategy, which last week observed, “It is becoming evident that the Chinese leadership is willing to abandon its financial de-risking agenda in exchange for a rapid economic recovery.”2 Our colleagues go on to note, “Monetary conditions are already more accommodative than during the last easing cycle in 2015/2016. The recently announced policy initiatives on infrastructure, housing, and automobile sectors also resemble policy supports that led to a V-shaped economic recovery in 2016.” Fed Cuts Rates To Reduce Uncertainty The economic pressure arising from a strong USD is particularly acute for EM economies. Even before the COVID-19 outbreak in China at the end of last year, global economic policy uncertainty (GEPU) and the broad trade-weighted USD (USD TWIB) were hitting new highs. This was driven by trade wars, the emergence of left- and right-wing populists globally, uncertainty over the effectiveness of monetary policy, and a host of other issues that drove investors, firms and households to seek safe-haven assets like the dollar (Chart 4). In fact, these variables became highly correlated over the past 3-, 4- and 5-year intervals.3 The novel coronavirus outbreak in China, which literally shut down China’s economy in January and February, added to this uncertainty. It continues to lurk in the background now that the coronavirus has spread globally. This also contributes to safe-haven USD demand. While a rate cut cannot address the COVID-19 directly, it can loosen financial conditions – thus removing some uncertainty at the margin – and reduce USD strength. The economic pressure arising from a strong USD is particularly acute for EM economies, which are the dominant source of commodity demand growth globally (Chart 5). At the margin, this demand for dollars arising from increased global policy uncertainty suppresses oil demand growth in EM economies, by raising its cost in local-currency terms ex-US and ex-GCC producing states with currencies pegged to the dollar. It also incentivizes production at the margin, as local-currency costs are depressed, which reduces local costs, while revenues are realized in USD – the perfect arb. Chart 4Global Uncertainty Was High Before COVID-19 Hit Markets Chart 5EM Growth Suppressed By Strong USD Exploring The Dominant Currency Paradigm The USD’s dominance of global trade is receiving considerable attention in academia and at the Fed. The USD’s dominance of global trade is receiving considerable attention in academia and at the Fed. One theory we find useful is the “Dominant Currency Paradigm,” which holds the dollar is the dominant currency in the world and is used disproportionally vis-à-vis its GDP weight in the global economy (Chart 6). Its dominance is reflected in (1) invoicing of international trade, (2) bank funding, (3) corporate borrowing, (4) central-bank reserve holdings, and (5) the relatively low expected returns accruing to USD-denominated risk-free assets that violate uncovered interest-rate parity no-arbitrage conditions – i.e., the dollar’s so-called “exorbitant privilege.”4 Chart 6USD Is The Dominant EM Invoicing Currency Demand for USD rises when global economic policy uncertainty rises, which is why dollar liquidity is crucial: When demand for safe asset spikes, there is a need for aggressive liquidity (supply) of dollar to avoid a market collapse (Chart 7).5 By cutting US rates now, the Fed is effectively increasing USD supply and/or removing some of the demand for USD relative to other currencies. This will be especially important if global economic policy uncertainty remains strong. This somewhat buffers EM corporates and governments with high levels of USD-denominated debt against a rush to safe-haven USD holdings. We believe this will ease financial conditions in EM economies, which should, all else equal, provide more of a shock absorber for uncertainty generally. Chart 7Dollar Liquidity Mutes US Dollar Appreciation   Our modeling suggests higher global economic policy uncertainty (GEPU) can shock the USD TWIB, US 10-year Treasurys and EM trade volumes directly. In addition, USD-denominated debt is relatively pronounced in some EM economies (Chart 8). USD appreciation increases domestic banks’ liabilities vs. assets. This is negative for bank’s balance sheets and leads to a tightening in financial conditions, which limits growth. EM corporate bond issuers are exquisitely sensitive to USD movements as they affect their capacity to service foreign-currency debt. Chart 8A Strong US Dollar Hurts Vulnerable EM Economies It is important to remember the US economy continues to perform relatively strongly against other major economies, with 1Q20 US GDP growth estimated by the Atlanta Fed’s Nowcast at 2.7% p.a. The fact that the Fed surprised markets with a 50bp rate cut suggests to us it is concerned with EM growth slowing sharply if the coronavirus becomes a global threat. The Fed also is likely to be concerned that lower US consumer confidence will lead to a decrease in the velocity of money. This concern also is addressed by increasing money supply pre-emptively. Our modeling suggests higher global economic policy uncertainty (GEPU) can shock the USD TWIB, US 10-year Treasurys and EM trade volumes directly, and that these shocks can persist (Chart 9).6 The Fed's policy action today will, if our modeling is correct, reduce demand for USD as a safe haven, all else equal, reduce long-term US rates and boost EM trade volumes. Bottom Line: We expect OPEC 2.0 to deliver at least 1mm b/d of production cuts in 2Q20, which will be reviewed at the end of June to determine whether they should be extended or deepened. Global economic policy uncertainty remains high, supporting demand for the USD. We believe the Fed’s surprise rate cut this week was directed at alleviating some of the global uncertainty keeping the USD well bid, in an attempt to buffer EM economies affected by USD demand. It also is a safeguard against a collapse in the velocity of money in the US that could occur if uncertainty were to suddenly rise. Chart 9GEPU Shocks Are Transmitted To USD And US Treasurys   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 The EIA’s weekly inventory report gives no evidence of a COVID-19-induced backup in crude and product inventories in the US. Total stocks of crude and products fell almost 12mm barrels last week on the back of strong product draws, led by gasoline and distillates, both of which were down close to 5mm barrels on the week. Commercial crude oil inventories were mostly unchanged at ~ 445mm barrels. Crude oil exports rose almost 500k b/d last week to 4.15mm b/d, accounting for most of the 9.73mm b/d of crude and product exports from the US. (Chart 10). Base Metals: Neutral Expectations China will deploy aggressive stimulus targeting infrastructure and manufacturing activities in response to the COVID-19 outbreak, along with Brazil reporting a 15% month-on-month decline in exports of iron ore helped iron ore and steel futures post significant gains earlier this week, with the Singapore Exchange's 62% Fe Iron Ore futures closing 5.6% higher on Monday. However, these gains were short-lived – and will remain capped in the short-term – as weak Chinese demand persists and steel rebar inventories remain at record highs. Precious Metals: Neutral Amid a broader market sell-off gold prices dipped 4.5% on Friday – the worst performance since 2013 – but have since recovered, on the back of the Fed’s surprise rate cut this week. The US central bank delivered an emergency 50bps rate cut on Tuesday, gold erased all the losses with spot prices rising 3.2% at the close, to reach $1645.27/oz. Silver followed a similar pattern rebounding 2.9% on Tuesday, closing at $17.22/oz. We are long both metals and believe more upside is yet to come if central banks around the world coordinate on additional monetary easing (Chart 11). Ags/Softs:  Underweight Expectations of a stronger stimulus in response to COVID-19 pushed soybeans higher for a third consecutive day on Tuesday, with prices hitting a 6-week high intraday. Bean prices then retreat and close 0.3% higher than the previous session. Gains were capped by favorable weather conditions in Brazil, leading analysts to expect a record harvest this season. Wheat also rebounded on the Fed’s rate-cut news after a sluggish week that saw prices falling almost 5%. Uncertainty still reigns though, as the Australian Bureau of Agriculture crop report predicts wheat output to recover 41% to 21.4 Mn Mt in 2020, due to rainfall ending a period of severe drought. The most active wheat futures were up 0.8% at Tuesday’s close. Chart 10US Crude Oil Exports Are Rising Chart 11Lower Real Rates Will Support Gold       Footnotes 1     We do not rule out the possibility KSA or the GCC core producers shoulder the lion’s share of the cuts they seek, in order to balance the market. 2     Please see China: Back To Its Old Economic Playbook? published by BCA Research’s China Investment Strategy February 26, 2020. It is available as cis.bcaresearch.com. 3    This heightened uncertainty – i.e., the increase in “unknown unknowns” markets are attempting to process – is a recurrent theme in our research. See, e.g., 2020 Key Views: Policy Uncertainty Continues To Drive Commodity Markets published December 19, 2019. It is available at ces.bcaresearch.com. 4    Please see Gopinath, Gita and Jeremy Stein. “Banking, Trade, and the Making of a Dominant Currency,” Working Paper currently under revision for the Quarterly Journal of Economics. 5    Gopinath (2016) finds that the dollar’s share as an invoicing currency for imported goods is approximately 4.7 times the share of U.S. goods in imports. Please see Gopinath, Gita. “The International Price System.” Jackson Hole Symposium Proceedings, published in January 2016. See also Obstfeld, Maurice (2019), “Global Dimensions of U.S. Monetary Policy,” presented at the Federal Reserve Board Conference on Monetary Policy Strategy, Tools, and Communication Practices (A Fed Listens Event) in Washington June 4, 2019. 6    Our results reflect the vector autoregression (VAR) model we use to study the interaction of GEPU shocks and the USD TWIB and US 10-year treasurys.   Investment Views and Themes Strategic Recommendations Tactical Trades TRADE RECOMMENDATION PERFORMANCE IN 2019 Q4 Commodity Prices and Plays Reference Table Trades Closed in 2020 Summary of Closed Trades

Feature “Bayesian: …statistical methods that assign probabilities or distributions to events…based on experience or best guesses before experimentation and data collection and that apply Bayes' theorem to revise the probabilities and distributions after obtaining experimental data.” — Merriam-Webster Dictionary Markets have reacted pretty rationally to the outbreak of the COVID-19 virus. Equities initially rebounded a few days ahead of the peak of new cases in China (Chart 1). But then, once the number of cases in the rest of the world started to accelerate, stock markets sold off again sharply. The MSCI All Country World Index is now down 13% from its peak on February 12. Recommended Allocation Chart 1Markets Have Reacted In Line With New COVID-19 Cases No one knows whether this episode will turn into an unprecedented pandemic, which will kill millions worldwide, last for months, and trigger a global recession. So it is the sort of environment in which Bayesian analysis becomes useful. Our “prior” for the probability of a full pandemic would be around 10-20%. If it doesn’t happen, an attractive buying opportunity for risk assets should present itself soon. But there could be further downside first, especially if the number of cases in major countries such as the US, Germany, and the UK were to accelerate significantly. There are some sign that Chinese activity is beginning to recover. There are some signs that Chinese activity is beginning to recover, as new cases of COVID-19 slow, thanks to the draconian measures taken by the authorities. Big Data can help analyze this. For example, live traffic statistics from TomTom show that by February 28, weekday road congestion in Shanghai was back to 50% of its normal level, compared to 19% on February 14 (Chart 2). The Chinese authorities have relaunched fiscal and monetary stimulus, causing short-term rates to fall to their lowest level since 2010 (Chart 3). Monetary policy has been upgraded from “prudent” to “flexible and moderate.” BCA Research’s China strategists believe there is even an increasing possibility of a stimulus overshoot in the next 6-12 months, as the authorities plan for the worst-case scenario but the economy rebounds.1 Chart 2Chinese People Getting Back On The Roads Chart 3Chinese Stimulus Pushing Down Rates In the short-term, it is clear that global growth will weaken, though quantifying this is hard. A 1% quarter-on-quarter decline in Chinese GDP in Q1 would bring growth down to 3.5% year-over-year. Our colleagues in BCA’s Global Investment Strategy estimate this would cause global growth to fall 0.8% below trend in Q1, mainly from a contraction in tourism, but that this would be largely made up in Q2, assuming that the epidemic is over by then (Chart 4).2 Could even a limited epidemic tip the world into recession? We doubt it. Consumer confidence remains strong in developed economies (Chart 5) and the virus is not yet serious enough to stop most consumers going out to spend. The global economy was in the process of bottoming out before COVID-19 hit (Chart 6) and there is little reason to think that we will not return to the status quo ante. Chart 4Global Growth To Slow In Q1, But Rebound In Q2   Chart 5Consumers Remain Confident Chart 6Before COVID-19, Growth Was Bottoming Out We see the two biggest risks being: 1) a rise in defaults in China, especially among smaller companies, that the government is unable or unwilling to prevent (Chart 7); and 2) a deterioration in the jobs market in the US, as companies start to postpone hiring, or lay off staff (Chart 8). We will watch these carefully over coming weeks. Chart 7Are Chinese Companies Vulnerable? Chart 8Is The US Job Market Starting To Wobble? Chart 9Markets Believe Trump Would Beat Sanders There is one other risk that might give equity markets an excuse for a further sell-off: November’s US presidential election. The probability that Bernie Sanders wins the Democratic nomination has risen to 60% from 15% over the past two months. The consensus believes that Trump can easily defeat Sanders, which is why the President’s probability of being reelected has risen in tandem (Chart 9). But, if the economy starts to weaken and Trump’s approval rating slips, investors could become nervous about the likelihood of a market-unfriendly Sanders administration. We would not recommend long-term investors sell out of risk assets at this point. There could be an attractive buying opportunity over the next few weeks, and investors who have derisked should be looking for a reentry point. With US 10-year bonds yields at 1.2% and German yields at -60 basis points, it is hard to see much further upside for risk-free bonds. Equities should be able to outperform over the next 12 months, as growth rebounds following the COVID-19 episode. We have been recommending overweights in cash and gold, as hedges, since December, and these still make sense. However, if events over the coming weeks point to the risk of global pandemic being higher than we currently think, then investors should Bayesianally adjust and move more risk-off. Otherwise, a peak in COVID-19 cases ex-China should be a strong signal to buy risk assets again. Chart 10Why Should Long-Run Inflation Expectations Fall? Fixed Income: US Treasurys have become investors’ safe haven of choice over the past few weeks. A marked drop in long-run inflation expectations (Chart 10), in particular, has pushed the 10-year yield to a record low. This seems somewhat illogical, since the Fed will announce this summer the results of its review of monetary policy, which is likely to lead to a more dovish long-term inflation target (perhaps a commitment to achieve 2% on average over the cycle). The market has also priced in at least three Fed rate cuts by year-end (Chart 11). The Fed will certainly cut rates if US growth falters as a result of COVID-19, but this is by no means a certainty. History shows that Treasury yields jumped sharply once previous viral outbreaks ended (Chart 12). We expect yields to be significantly higher in 12 months, and so are underweight duration and prefer TIPS over nominal bonds. Credit will continue to underperform in the risk-off phase, but some interesting opportunities should arise soon, especially among the lowest-rated credits and in the Energy sector. Chart 11Will The Fed Really Be This Accommodating? Chart 12After Previous Virus Outbreaks, Rates Leapt Equities: The sell-off has already put on fire sale some stocks most affected by the epidemic. For example, cruise lines are down by 40% over the past month or so, European oil stocks 25%, some luxury goods makers 30%, and airlines 30%. Opportunistic investors might want to buy a basket of the most oversold quality names. Our overweight on euro area stocks has not worked in the sell-off. But, as a cyclical, export-oriented market, we continue to expect Europe to outperform when global growth rebounds. Euro area banks, in particular, represent the best call option on a rise in bond yields, since their performance is highly correlated to the shape of the yield curve. We continue to have a somewhat cyclical tilt among our sector weightings (with overweights on, for example, Energy and Industrials), but may adjust this in our Quarterly Portfolio Outlook in early April if we decide to reduce risk. The sell-off has already put on fire sale some stocks most affected by the epidemic. Currencies: The dollar is a safe-haven currency and so, unsurprisingly, has benefitted from the rush to safety in recent weeks. However, it remains overvalued (Chart 13), and interest rate differentials would move further against it if the Fed does cut rates, since other major developed central banks have much less room to move (Chart 14). This suggests that it will probably resume the weakness it experienced from August to December last year as soon as global growth rebounds. Chart 13Dollar Is Overvalued... Chart 14...And Interest Differentials Have Moved Against It Chart 15Metals Prices Stabilized In Recent Weeks Commodities: Industrial metals fell sharply on the outbreak of COVID-19 in China, but have bottomed in line with the stabilization of the situation in that country (Chart 15). Gold has worked predictably as the best hedge in the sell-off. While it is starting to look technically overbought and would be hurt by a rise in bond yields (Chart 16), for prudent investors it remains a useful hiding place amid heightened risk and ultra-low interest rates. Oil is the commodity that has fallen the most surprisingly, with Brent close to the low it reached during the sell-off in December 2018 (Chart 17).  It is much less dependent on Chinese demand than metals are, and so is maybe pricing in a global recession – as well as questioning the commitment of OPEC to cut production further. This would suggest upside to the oil price if global growth turns out not to be so bad, oil demand continues to pick up, and supply remains constrained.   Chart 16How Much Could Gold Overshoot? Chart 17Oil Discounting A Global Recession Garry Evans, Senior Vice President Chief Global Asset Allocation Strategist garry@bcaresearch.com Footnotes 1    Please see China Investment Strategy Weekly Report, “China: Back To Its Old Economic Playbook?” dated 26 February 2020, available at cis.bcaresearch.com 2   Please see Global Investment Strategy Weekly Report, “Market Too Complacent About The Coronavirus,” dated 21 February 2020, available at cis.bcaresearch.com GAA Asset Allocation  

Highlights Supply constraints and unstoppable demand growth – the result of stricter regulations requiring higher loadings in autocatalysts to treat toxic pollution in automobile-engine emissions – will continue to push palladium’s price higher, despite a near-vertical move higher that began in 2H19. South Africa’s power grid is in a state of near-collapse, which will add volatility to mining operations focused on platinum-group metals – chiefly palladium, platinum and rhodium. South Africa accounts for 36% of global palladium production and 73% of platinum production, which makes it difficult to make the case that platinum could be substituted for palladium as its price rises. Palladium stocks are at risk of being further depleted globally as demand from automobile manufacturers in China, the US and Europe remains robust. This will keep palladium forward curves backwardated for the foreseeable future. While pressure to find alternatives for palladium will grow as prices rise, in absolute terms the additional cost resulting from higher prices for the metal – ~ $400 per vehicle – is not yet enough to draw significant investment to this effort. Feature Palladium markets are fundamentally tight and unresponsive to macroeconomic uncertainty. Table 1Top 5 Best Performing Commodities In 2019, for the third year in a row, palladium prices outperformed other major commodities, returning an impressive 54% over the year (Table 1). This is the result of a massive 13% increase in demand for the metal – powered by strong autocatalyst demand for gasoline-powered cars in China and Europe, even as collapsing auto production globally and elevated trade uncertainty continue to dog automobile sales (Chart 1). This apparent contradiction is explained by stricter vehicle emissions regulations in major consuming markets – chiefly the Euro 6d, China 6 and US Tier 3 regimes – and power shortages in South Africa, which are introducing considerable volatility on the supply side in the second-largest producing country for the metal. Chart of the WeekSurging Autocatalyst Palladium Demand  Again this year, palladium markets are fundamentally tight and unresponsive to macroeconomic uncertainty. Palladium prices soared 39% YTD, its fastest 40-day increase since 2010. Unlike other commodity markets, palladium is completely disregarding the COVID-19 outbreak that originated in China late last year. Favorable supply-side fundamentals continue to drive the palladium rally: The metal’s decade-long physical supply deficit intensified in 2019 and we expect it to widen this year (Chart 2, panel 1). On the demand side, Chinese consumption is at risk. China is the world’s largest auto manufacturing market. Hubei Province – COVID-19’s epicenter – is a large car manufacturing hub, accounting for ~ 10% of the country’s annual automobile output. In the wake of COVID-19, the country’s car production is expected to fall 10% in 1Q20. In addition, the virus had infected more than 80,000 people globally, and has spread rapidly outside Hubei into Asia, Europe, the Middle East, Africa, and North America, raising the odds of a pandemic. Interestingly, speculative positioning and ETF investment demand is subdued, and is not inflating prices (Chart 2, panel 2). Chart 2Palladium Deficit To Widen This Year Palladium Demand Soars As Auto Production Collapses Strong global automobile catalyst demand drove the rally in palladium prices last year. This occurred as car production fell by 9%, 8%, and 15% in US, China, and India – an unusual divergence in fundamentals. The culprit: Technical changes to autocatalysts from stricter emissions regulations. In China, the latest phase of car emissions regulations – China 6 – was gradually introduced in high-population centers, which also suffer from high levels of pollution. These centers accounted for ~ 60% of annual Chinese car sales in 2019. China 6 represents a major shift in emissions regulations and will make the Chinese auto fleet compliant with Europe’s best practices. As a result, palladium loadings in conforming light-duty gasoline vehicles reportedly increased by ~20% in 2019. This pushed China’s autocatalyst consumption up by 570k oz despite the drop in annual car sales, which created the rare dislocation between the country’s car production and palladium prices (Chart 3). We expect this trend to continue this year: China 6 is on track to be enforced countrywide – i.e., the remaining 40% of car sales – by mid-year, providing an additional ~ 10% boost in loadings of the metal. Chart 3Stricter Regulations Support Prices Amid Falling Car Production In Europe, the introduction of Euro 6c legislation in September 2018 and the extension to all new vehicles of Euro 6d-TEMP regulations in September 2019 – mainly the real driving emissions (RDE) testing procedure adopted in the wake of the Volkswagen “dieselgate” scandal in 2015 – pushed palladium loading in autocatalysts up by ~ 25% from 2017 to 2019.1 The regulations became stricter in January 2020, putting additional stress on manufacturers to comply with the new standards, which will continue to support higher palladium loadings. We expect the COVID-19 outbreak to delay the recovery in global gasoline-powered vehicle production and consumption to 2H20. Lastly, in the US – which remains an important market for autocatalyst palladium demand (Chart 4) – the ongoing implementation of the Tier 3 legislation will continue to gradually increase palladium content in autocatalysts until 2025. For 2020, we do not expect this to significantly boost loadings per vehicle and are factoring in 2% growth. These legislative changes in major automotive markets produced a structural break in our palladium demand model (Chart 5). After adjusting our estimates for greater palladium content in gasoline aftertreatment systems, our model suggests that demand provides strong support to palladium prices, but also suggests other factors – i.e. supply and inventory – are at play. Chart 4North America's Auto Sector Remains A Large Share Of Palladium Demand Chart 5Higher Palladium Loadings Largely Explains Last Year's Price Surge In the US and Europe, consumers can absorb higher vehicle sales despite being close to saturated in terms of vehicle ownership. We expect the COVID-19 outbreak to delay the recovery in global gasoline-powered vehicle production and consumption to 2H20. In China, we expect the government will overstimulate its economy to meet its long-term goal of doubling its GDP and per capita income by 2020.2 Automobile ownership and vehicle sales there are low vs. DM economies, suggesting more upside for sales in China (Chart 6). In the US and Europe, consumers can absorb higher vehicle sales despite being close to saturated in terms of vehicle ownership. Car sales move in cycles around long-term demographic trends: The longer the current economic expansion, the further above-trend car sales can rise (Chart 7). Chart 6China: Structural Outlook For Autos Is Bright China Car Consumption Will Rebound In 2H20... Chart 7... Likewise For Europe And US Bottom Line: The combination of stricter environmental regulations in key gasoline-powered automobile markets and the post-coronavirus rebound in global auto consumption will push the palladium market further in deficit this year as it faces an inelastic supply, critically low inventories and low substitutability over the short-term (more on this below). Palladium Supply In 2020: Weak growth And Low Price-Elasticity Palladium supply is highly constrained. The largest supplies are concentrated in Russia (42%), South Africa (36%) and North America (14%). From 2015 to 2019, supply and capex grew by a very subdued 7% and 15.2% respectively, completely disregarding the 200% rise in prices (Chart 8, panel 1). This illustrates palladium supply’s extremely low price-elasticity.3 Palladium supply growth will remain muted for the foreseeable future, as Eskom begins long-delayed maintenance to refurbish its derelict generation fleet. Primary supplies declined by close to 2% last year on falling shipments from Russia and record electricity load-shedding – i.e. blackouts – in South Africa (Chart 8, panel 2).4 As tight as palladium markets are fundamentally, South Africa’s crippled power grid – long in need of upgrading and repair – has been, and remains, a key driver of short-term platinum-group metals (PGM) prices.5 Following the breakdown of close to 25% of the country’s generating capacity, Eskom – the nation’s utility monopoly responsible for ~ 90% of its electricity generation – has been forced to implement rolling blackouts to balance power supply and demand and prevent permanent damage to the country’s power grid. Palladium supply growth will remain muted for the foreseeable future, as Eskom begins long-delayed maintenance to refurbish its derelict generation fleet. Consequently, Stage 6 load-shedding events likely will become more frequent. These efforts are complicated by massive debt – ~ $30 billion – which has required government bailouts and forced the company to take loans from a Chinese industrial bank. Chart 8Top Palladium Producers' Capex Price-Elasticity Is Low This is playing havoc with PGM supplies. During the unmatched Stage 6 load-shedding in December 2019 – cutting power to 37% of grid users – PGM supplies were reduced by 50%. Stockpiles covered the loss, but persistent blackouts lasting years could push markets into an actual shortage of palladium as inventories would rapidly be depleted. This is a significant risk: Eskom itself warned rolling blackouts will persist for the next 18 months.6 Elevated local currency PGM prices are postponing announced shafts closures, as miners seek to profit from the favorable pricing environment (Chart 9). But insufficient electricity capacity will weigh on mine supply growth over the next few years as companies hold-back on much-needed long-term investments. The final units of Eskom’s Medupi and Kusile projects are expected to be completed over the next two years – adding 4800MW to its installed capacity. This can partially alleviate South Africa’s electricity difficulties, but these units are not enough to support a rebound in economic and mine production growth. South Africa is in profound need of large-scale investments in its power sector. Close to 5000MW of power capacity is scheduled to shut down over the next five years (Chart 10). Chart 9Favorable Domestic Metal Prices For South African Miners Chart 10South Africa Needs Additional Power Generation Capacity After years of pressure from mining companies, South Africa’s minister of Mineral Resources and Energy announced it would allow companies to generate unlimited electricity for their own activities. The current political and economic climate is not constructive for meeting this challenge. The World Bank recently slashed South Africa’s 2020 GDP growth forecast to 0.9% from 1.5% previously on the back of electricity and infrastructure constraints impeding domestic growth and weak external demand. Likewise, rating agency Moody's signaled – ahead of its review of South Africa’s Baa3 credit rating in March – it could downgrade the country to speculative grade, citing the detrimental impact of recurring power outages on manufacturing and mining output. After years of pressure from mining companies, South Africa’s minister of Mineral Resources and Energy announced it would allow companies to generate unlimited electricity for their own activities. This will provide much-needed help to the country’s power sector. According to the Minerals Council South Africa, mining companies could bring an additional ~ 1500MW capacity online in the next 9 to 36 months. But doubts remain with regard to the timeline for companies to obtain the necessary licenses and if these can easily be acquired. Johnson Matthey expects supply growth in Russia – the largest producer – will be capped this year as Nornickel’s processing of old mines' copper concentrate – which boosted the company’s palladium supply over the past few years – is finalized. Still, a paltry 1% gain is possible from expected efficiency gains at existing mines, according to Nornickel. The company also announced it will increase production at its Talnakh and South Cluster mines, but this additional supply will only reach markets gradually as processing capacity constraints won’t be resolved until 2023, according to Johnson Matthey. Bottom Line: Growth prospects in the top two palladium-producing countries are weak in 2020. This will not suffice to meet the soaring autocatalyst demand. Higher recycling and inventory releases – both incentivized by higher prices – will be needed to balance the market. Palladium Stockpiles Are Dangerously Low We expect palladium prices will move higher on the expanding deficit, and backwardation in the forward curve will persist to incentivize the release of inventories to market (Chart 11). Yet, global palladium stockpiles have been declining since 2014 and are now at critically low levels, raising the risk of a disrupting shortage of the metal:7 ETF and exchange inventories now stand at a paltry 600k oz (Chart 12). These are the most price-elastic stocks and will get close to zero as prices increase. Chart 10Expect Backwardation To Persist Chart 12Price-Sensitive Stockpiles Are Dangerously Low Exhaustion of inventory would spike prices until demand destruction or additional supply – both inelastic in the short-run – are able to balance the market. The Russian Ministry of Finance’s reserves – a state secret – are now almost exhausted, according to Russia’s Norilsk Nickel, the largest supplier of physical palladium in the world. Last year, Norilsk Nickel held an estimated 1mm oz of the metal in its Global Palladium Fund, and signaled it is increasingly using its reserves to balance markets and provide needed liquidity. Earlier this year, the company released 3 MT of palladium to the market from stocks. Complete exhaustion of inventory would spike prices until demand destruction or additional supply – both inelastic in the short-run – are able to balance the market. Don’t Count On Substitution, Yet Switching to platinum requires significant capital- and resource-intensive R&D and appears to be beyond the current capabilities of automakers. We expect platinum prices to rise in 2020 supported by improving fundamentals, growing safe-haven demand, and markets pricing in increasing anticipation of substitution from palladium to platinum. Unlike palladium, platinum is also affected by safe-haven demand and gets bid up with gold and silver prices in periods of high uncertainty (Chart 13). With gold prices now above $1,600/oz, platinum will benefit from safe-haven flows due to its relative price advantage (Chart 14). Chart 13Safe-Haven Flows Support Platinum Prices Chart 14Platinum Is Cheap Relative To Gold We believe substitution will commence over the coming years, but this is a gradual process. Substitution from expensive palladium to low-priced platinum in industrial applications is the largest risk to our positive view on the palladium-to-platinum (Pd-to-Pt) ratio (Chart 15). This started in smaller and more price-elastic segments (e.g. dental, jewelry and diesel autocatalyst). However, to have a real impact on overall demand and thus the price ratio, substitution needs to take place in gasoline autocatalyst technology. The discount has been at a level consistent with substitution for more than a year, but the urgency to upgrade current designs to meet new environmental legislation and RDE regulations in China, Europe, and the US is the main focus of automakers this year. Switching to platinum requires significant capital- and resource-intensive R&D and appears to be beyond the current capabilities of automakers scrambling to meet the latest anti-pollution regulations globally. Moreover, large-scale substitution will take place only if automakers’ cost-benefit analysis points to significant long-term profits from switching. That said, platinum’s supply security remains a risk in the long-term: South Africa accounts for 73% of global production and our analysis suggests output growth there likely will remain weak over the next few years, especially as Eskom rebuilds its failing power grid. This lack of diversity increases sourcing risks for automakers, who, not without reason, would not want to switch over to platinum only to find that supply is also in doubt down the road. The overall platinum market is 26% smaller than that of palladium. Assuming a one-for-one substitution of Pd to Pt in gasoline catalyzers, a 1.2mm oz reduction in Pd demand – the amount required to reduce palladium’s deficit to zero – would send platinum markets to a 1.4mm oz deficit.8 Without substantial production growth, platinum prices would spike, reducing the profitability of investing in these new catalysts. Thus, substitution will eventually impact the price ratio, but will not be large enough to overturn absolute price level trends. In addition, the amount of PGMs in the typical autocatalyst – ~ 5 grams – adds $400 to the cost of the average automobile (Chart 15, lower panel). We do not believe this cost drives automakers' decisions, which is another reason the substitution of Pt for Pd likely will remain a topic of discussion more than action. Chart 15Palladium's Price Surge Adds ~0 Per Gasoline Car Bottom Line: We believe substitution will commence over the coming years, but this is a gradual process and it will not happen on a meaningful scale this year. Thus, we expect the continuation of relative demand and inventory trends will provide a favorable setting for the Pd-to-Pt ratio this year (Chart 16). Chart 16Pd-to-Pt Price Ratio Will Increase Again in 2020   Hugo Bélanger Associate Editor Commodity & Energy Strategy HugoB@bcaresearch.com Robert P. Ryan Chief Commodity & Energy Strategist rryan@bcaresearch.com   Commodities Round-Up Energy: Overweight Brent and WTI crude oil lost 5% and 4% this week, as fears of a global pandemic in the wake of the COVID-19 outbreak gripped markets. Reports of outbreaks in Asia ex-China, the Middle East and Europe fueled these concerns. Against this backdrop, OPEC 2.0 will be meeting in Vienna March 5 and 6 to consider cuts of 600k b/d recommended by its technical committee earlier this month. We continue to expect the full coalition to approve these cuts at the upcoming meetings. Saudi Arabia, Kuwait and the United Arab Emirates reportedly are considering an additional 300k b/d of cuts to offset the global demand hit delivered by COVID-19. The IEA estimates the COVID-19 outbreak will reduce Chinese refining throughput by 1.1mm b/d, and will reduce the call on OPEC crude by 1.7mm b/d in 1Q20. Base Metals: Neutral Iron ore prices weakened, following global equities lower, as the COVID-19 outbreak spread around the world. However, traders continue to report lower stocks of iron ore, which should keep prices supported, according to MB Fastmarkets (Chart 17). We remain long December 2020 high-grade iron ore (65% Fe) vs. short the benchmark 62% Fe contract on the Singapore Commodity Exchange, which we initiated November 7, 2019. This recommendation was up 5.3% as of Tuesday’s close, when we mark to market. Precious Metals: Neutral After retreating slightly from its run toward $1,700/oz earlier this week, gold remains well supported by safe-haven demand (Chart 18).  In addition, actual and expected policy stimulus – e.g., Hong Kong's “helicopter money” drop of USD 1,200 to all permanent residents over the age of 18 – and expectations of additional central bank easing globally to offset the global spread of COVID0-19 will keep gold and precious metals generally supported.  Markets should start pricing in higher inflation expectations as additional stimulus starts to roll in.  Ags/Softs:  Underweight Global grain markets could be set to rally sharply, as unusually wet weather in the Middle East and East Africa spawned by higher-than-usual cyclone activity produces perfect breeding conditions for desert locusts in the region over the next two months.  According to National Geographic, by June the locusts could increase their populations “400-fold compared with today, triggering widespread devastation to crops and pastures in a region that’s already extremely vulnerable to famine.”  This could put more than 13mm people in East Africa at risk of “severe acute food insecurity,” and imperil millions more.  Chart 17China's Iron Ore Stocks Tight Chart 18Safe Havens Gold, USD Well Bid   Footnotes 1     Please see New legislation planned in response to dieselgate, published by Autocar June 9, 2016. See also  Johnson Matthey’s February 2020 Pgm Market Report. 2     Our view of strong Chinese fiscal and monetary stimulus was discussed in detail in our February 13, 2020 weekly report titled Iron Ore, Steel Poised For Rally. 3    Historically produced as an inferior byproduct from nickel, gold, and platinum mines, the price incentive from palladium alone isn’t enough to generate the needed investments in new mine production. According to Nornickel, this is slowly changing, palladium is an increasingly large part of mining companies’ revenues, making the metal a valuable co-product. This could improve mines investments’ responsiveness to movement in palladium prices over the medium term. 4    According to Eskom, “Load shedding is aimed at removing load from the power system when there is an imbalance between the electricity available and the demand for electricity. If we did not shed load, then the whole national power system would switch off and no one would have electricity.” The company’s load-shedding program includes 8 stages, where each stage represents the removal of 1000MW of demand – e.g., stage 5 removes 5000MW. This is done by shutting down specific sections of the grid.  5    The PGMs are ruthenium, rhodium, palladium, osmium, iridium, and platinum. 6    Things got worse after the December load-shedding event.  Less than a month later, Reuters noted more than two times the power shed in December went “offline because of plant breakdowns. 7    This can be seen in the close to 12mm oz. decline in UK and Switzerland – home of the largest secured vaults of Palladium and Platinum – net imports. 8    Technological improvement in palladium catalysts has made the metal more efficient in for gasoline-powered engines vs. platinum. It has superior properties in terms of thermal durability and NOx reduction. Thus, the conversion could be greater than 1-to-1 and would imply a smaller share of palladium autocatalyst substitution could be absorbed by existing platinum supplies.   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

Special Report

Highlights Investors’ hunt for yield over the past few years has increasingly led them to view emerging markets debt (EMD) as an attractive component of portfolios. EMD should not be viewed as one homogeneous asset class. Investors should distinguish between its key segments: hard-currency sovereign debt, hard-currency corporate debt, and local-currency sovereign debt. EMD allows investors to own bonds with higher yields than DM sovereign or corporate bonds. But it comes with specific risks that investors need to understand. EMD, being a highly cyclical asset class, should perform well in an environment of accelerating global growth – which we expect to see during 2020. Within this asset class, we favor EM hard-currency sovereign bonds over both EM hard-currency corporate debt and local-currency sovereign bonds. However, the coronavirus outbreak makes us reluctant to pull the trigger on this recommendation now. Rather, we are placing EM hard-currency sovereign debt on upgrade watch. Feature Emerging markets debt (EMD) as an asset class has grown over the past decades to over US$24 trillion in bonds outstanding – becoming an integral part of the global investment universe, and presenting an interesting investment opportunity for investors. The EMD universe, which was previously dominated by sovereign issues in hard currencies, has become more diverse, and consequently, difficult for investors to ignore. In this Special Report, we identify the segments that make up EMD and the various exposures that investors face when allocating to it. We analyze their risk-return characteristics and the drivers contributing to their returns, and compare EMD to other asset classes. We conclude by identifying any diversification benefits that investors can reap as the hunt for yield continues. Introduction Estimates value total debt in emerging markets at over $24 trillion as of Q2 2019. This includes both sovereign and corporate debt, in both local and hard currencies (Chart 1).1  The bulk of EMD, however, is in local currencies – almost 90%. Chart 1Estimates Of Total EMD In this Special Report, we focus on the following three segments of emerging markets debt (EMD): Sovereign debt issued in hard currency – the majority of which is USD denominated – estimated at $2 trillion. I. We distinguish between “pure”            sovereigns and quasi-sovereign bonds. Corporate debt2  issued in hard currency – mainly in USD – estimated to be $1.5 trillion. Sovereign debt issued in local currency – estimated at $10.3 trillion. We do not cover local-currency corporate debt, as more than half of it – estimated to be $8.1 trillion – is issued by Chinese firms and is hard to access for most investors. Each of these segments offers an array of opportunities, is driven by different dynamics, and bears risks that investors must recognize before allocating to it. We recommend clients view the segments of EMD as different asset classes, rather than an aggregate. Hard-Currency Debt Hard-currency EMD refers to debt issued by governments and firms in emerging markets that is denominated in a currency other than their local currency. Estimates suggest 90%-95% of total hard-currency debt is USD denominated, with the remaining in euros and yen. The main feature of hard-currency EMD is that it provides investors with protection against currency depreciation risk. Nevertheless, it is important to highlight that currency movements can affect spreads, default risk, as well as liquidity. If a country’s currency depreciates, its ability to service its foreign debt deteriorates. This is crucial, as exemplified by currency crises over the past few years in countries such as Argentina, Turkey, Egypt, and Venezuela. Hard-Currency Sovereign Debt Since 2004, EM hard-currency sovereign bond investors have enjoyed an annualized total return of 7.4%, much higher than the 3.2% from the global Treasury index. Even on a risk-adjusted return basis, the incremental performance compensates for the additional 1.7% of annualized volatility. Investing in EM hard-currency sovereigns allows investors to find higher-yielding debt than government bonds in developed economies. Since 2004, the average yield on EM hard-currency sovereign debt was 6.1%, 3.8 percentage points higher than the 2.3% on their DM counterparts. Investors received positive returns even in real terms, as inflation in DM and the US have averaged 2.2% and 2.1% respectively, since 2004 (Chart 2). This has been extremely useful, particularly in the past few years, when bond yields in many developed economies reached zero or turned negative, and investors increasingly hunted for yield. The risk profile of the aggregate EM sovereign debt index is balanced between the safer Middle Eastern economies such as Saudi Arabia, UAE, and Qatar, and the riskier Latin American economies such as Mexico, Brazil, and Argentina. Those two buckets each comprise approximately 30% of the index, with the remainder of the index split between Asia, Emerging Europe, and Africa at 17%, 11%, and 10%, respectively (Chart 3). Other portfolios are benchmarked to J.P. Morgan’s indexes where Gulf countries have very little weight. Chart 2EM USD-Sovereigns Provide Value To DM Investors Chart 3Risk Profile Of EM USD-Sovereigns We believe it is reasonable to compare hard-currency EM sovereign debt to US investment-grade bonds due to their shared characteristics. Both have comparable duration (approximately eight years) and similar credit qualities, despite EM sovereign debt being a little riskier on average than the US corporate market (Chart 4). Nevertheless, since 2004, EM sovereign hard-currency debt has outperformed US investment-grade bonds by 40% – although its outperformance has lost steam over the past few years (Chart 5). Chart 4EM USD-Sovereigns Are Slightly Riskier Than US Investment-Grade Bonds Chart 5EM USD-Sovereigns Have Outperformed US IG Bonds... This does not mean that EM debt is immune to problems.3 The cumulative average default rate of EM foreign-currency sovereign debt – while lower than US corporates – remains high and is more pronounced as one goes down the credit-rating curve (Table 1). Idiosyncratic country risks can skew the data. If one excludes Argentina – currently weighted at only 3.5% – from the index, almost 100 basis points of spread get shaved off (Chart 6). Table 1…However, Beware Of The Default Rates Chart 6Excluding Argentina, Spreads Are Much Lower Given that most of our clients invest through passive vehicles, throughout this report, we focus on the EM aggregate indexes rather than on specific countries. However, it is important to identify over/undervalued countries, given the wide-ranging risk-profile spectrum of emerging economies. By drawing a US corporate credit curve, based on credit ratings and breakeven spreads, one can spot over- or undervalued countries relative to US investment-grade bonds. Currently, the sovereign bonds of Poland, UAE, Qatar, and Saudi Arabia appear to be more attractively valued than those of Russia, Hungary, and Brazil. The charts also show the transition of these countries across time (Chart 7, A,B,C,D). Chart 7Country Selection Is Important… Chart 8...With The UAE And Saudi As Good Examples For example, South African sovereign bonds – given their current credit rating and spreads – have moved from being overvalued relative to US corporates to undervalued over the past five years. This implies a buying opportunity, or simply that they are getting cheaper ahead of a potential downgrade. For investors with less restricted mandates, country selection can be very valuable. For example, the UAE and Saudi Arabia, two highly rated economies at Aa2 and A1 respectively, trade at 23 and 30 basis points over similarly rated US corporate bonds (Chart 8). We find that EM hard-currency sovereign spreads are mainly driven by global growth cycles, something BCA Research’s Emerging Market strategists have often highlighted.4  We rely on several key indicators to gauge where we are in the cycle. These include Germany’s IFO manufacturing business expectations, global and emerging market PMIs, as well as OECD’s Leading Economic Indicators (LEI) (Chart 9). Upward moves in these indicators have historically led to a tightening in EM sovereign spreads. Chart 9Spreads Will Tighten Once Global Growth Picks Up Quasi-Sovereign Bonds Chart 10Quasi-Sovereigns Are Focused In The Energy Sector Investors need to differentiate between EM sovereign bonds and quasi-sovereign bonds. While formal definitions vary among market participants and academics, the most common definition of “quasi-sovereign” is bonds issued by an entity where the government either fully owns the institution, controls more than 50% of its equity, or has a majority of its voting rights.5 Examples of such companies include Brazil’s Petrobras, Mexico’s Pemex, and Venezuela’s PDVSA. One reason why we highlight quasi-sovereigns is the rapid growth in the amount of such debt outstanding.6 As of January 2020, the quasi-sovereign bond market has grown by over US$630 billion throughout the past decade to US$714 billion and it now makes up over 42% of the combined EM Sovereign amd Quasi-Sovereign Bloomberg Barclays index. The oil & gas sector represents over a third of quasi-sovereign entities (Chart 10). Chart 11Quasi-Sovereigns...A Defensive Play On Corporate Bonds Some investors assume that a quasi-sovereign entity would have the full backing of its government. While that is true in most cases, the majority of quasi-sovereign bonds only have an “implicit” backing from the issuer’s government, meaning that the government holds no legal liability in case of default. Dubai World, a state-owned conglomerate, was a perfect example of this during the aftermath of the Global Financial Crisis. The government stood on the sidelines as the firm went through financial distress, forcing billions of dollars of debt to be restructured.7 Given this additional level of uncertainty and corporate risk, EM quasi-sovereign bonds trade at higher spreads than their sovereign counterparts (Chart 11). Nonetheless, bonds with even the simplest implicit backing from the government are considered a more defensive play than “pure” corporate bonds, which trade at even higher spreads. Hard-Currency Corporate Debt The increase in quasi-sovereign issuance has been a big factor in the growth of the hard-currency corporate-debt universe – a segment that became of interest to investors in the early 2000s. The outstanding amount of hard-currency corporate debt has surpassed hard-currency sovereign debt, according to the Bank Of International Settlements (BIS) (Chart 1). The EM corporate debt index8 has similar sector exposure to the MSCI EM equity index. Almost 69% of the bond index is concentrated in the Industrials category,9 with the Financial/Banking and Utilities sectors making up the remaining 26% and 5%, respectively (Chart 12). The Technology sector is an exception – comprising only 5% of the corporate bond index compared to over 16% in the equity index. The country exposure, however, is less skewed to Asian economies compared to equities (Chart 13). Chart 12EM Corporates Provide Similar Sector Exposure To Equities… Chart 13…Yet With Different Country Exposure Chart 14EM Corporates: A Defensive Play On Equities...   The overlap in sector coverage can be advantageous to investors who want quasi-exposure to EM equities but with much lower volatility. The same can be said for DM corporate bonds, whose return is highly correlated to equities but with about one-third the beta.10 The correlation between EM corporate bonds and EM equities is currently close to its post-2003 average of 0.61, and the beta of EM corporate bonds to EM equities has averaged only 0.13 (Chart 14). Despite having a lower annualized return11  than EM equities, 5.6% versus 8.3%, EM corporate bonds had almost half the realized volatility, and so outperformed equities on a risk-adjusted basis. In fact, since late 2007, they have generally outperformed EM equities even in absolute terms, despite a few periods of EM equity outperformance. Like sovereigns, EM corporate bonds provided investors with a cushion against equity downside risk. For example, during the 2015/2016 slowdown in China and emerging economies, EM equities fell by almost 28%, whereas EM corporate bonds fell by only 5% (Chart 15). Chart 15...With Lower Drawdowns On a valuation basis, however, EM corporate bonds have looked unattractive relative to EM equities, providing investors with 4% real yield, compared to an equity earnings yield of 7% since 2004 (Chart 16). Nevertheless, the current level of spreads points to moderate returns for the asset class, slightly below 4% annualized over the next five years, assuming that historical default and recovery rates remain the same, and no change in spreads (Chart 17). This implies that exposure to emerging markets via corporate bonds should be more attractive than equities on a risk-adjusted basis.12 Chart 16EM Corporate Bonds Are Unattractive Compared To Equities Chart 17Forward Returns Driven By The Spread EM corporate debt is similar to its sovereign counterpart in the range of risk profiles of its constituents. Default figures vary significantly by region and during different crises. For example, the hard-currency corporate default rate for Argentinian corporates peaked at slightly over 50% during the 2001/2002 sovereign debt crisis, while for Chile and Mexico it remained below 10%. Surprisingly, default rates in emerging market corporate speculative-grade debt have on average been below those of both the US and Europe (Chart 18). Additionally, the 12-month trailing default rate for the overall EM corporate universe, as measured by Moodys’ Investors Service, at the end of 2018 was lower than for advanced economies – at 1.4% versus 1.6%.13 Chart 18Default Rates In EM Are Surprisingly Lower Than In DM Chart 19EM Corporates Suffer From Weaker Balance Sheets   EM corporate spreads are driven by a few main variables – revenue and profit growth, the business cycle, and the exchange rate. The health of EM corporates is also an important factor. This is an area of concern as corporate leverage levels have risen since 2010, and EM firms’ ability to service debt – gauged by their interest-coverage ratio – has fallen to below 2008 levels (Chart 19). Political turmoil can upset markets. Even though investors do not face the risk of currency depreciation with hard-currency debt, EM corporates with revenues mostly in local currency, face higher debt-repayment risk during a slowdown in their economies. Local-Currency Sovereign Debt Chart 20There Is Value In EM Local-Currency Bonds Emerging-market governments, to avoid foreign currency liquidity crunches, have in recent years shifted some of their debt issuance to their own currency. However, to attract investors, yields on local-currency sovereign bonds have to compensate for the added layer of currency risk as well as conventional sovereign risk. Over the lifetime of the index,14 since 2003, yields on local-currency sovereign debt have averaged 6.7%, compared to 2.5% for the US Treasury index, 2.4% for the euro area treasury index, and 0.63% for the Japanese treasury index (Chart 20). Since 2004, EM local sovereign bonds have provided investors with attractive returns. On an annualized basis, they have returned 8.4% and 6.8% in local terms and dollar terms, respectively, albeit with higher volatility than their hard-currency counterparts on a common-currency basis (Table 2). However, those returns remain higher than those of government bonds in developed economies such as Germany and Japan, both in local- currency terms and on an unhedged basis from a USD perspective. Table 2EM Local-Currency Bonds Outperforming Other DM Government Bonds In USD And Local Currency Terms Investors allocating to this segment assume a simple yet plausible notion: that EM economies will never default on debt issued in their own currency, as they can easily “print more money”. This is partially correct: default rates across rated EM sovereign local debt remain lower than for foreign-currency sovereign debt (Table 3). Table 3Default Rates: Local-Currency Debt Versus Hard-Currency Debt Most interestingly, the gap in default rates between B- and CCC-rated bonds illustrates the “near certainty” of default for low-credit-rated sovereigns ahead of time. However, proponents of the notion that governments will not default neglect the consequences those economies will suffer if they monetize public debt: currency devaluation and high inflation, which turn into weak economic growth and tightening monetary policy, leading to a further weakening in growth.  The case of Argentina between 1998 and 2002 is a perfect example of this mechanism. The economy was hurting under an uncompetitive pegged currency as well as a large debt burden. The government’s move to increase taxes, as a solution to boost government revenues, triggered a cascade of events which resulted in faltering economic growth, increased unemployment, abandonment of the currency peg, and interest rates as high as 100%, ultimately leading to Argentina’s default on its local-currency sovereign debt (Chart 21). Chart 21Argentina: A Case Study Chart 22Country Breakdown Of Local-Currency EM Sovereigns Argentina was recently removed from J.P. Morgan’s EM local-currency sovereign index due to the capital controls the authorities have instituted. As of mid-February, Mexico was the largest issuer in the index along with Indonesia, Brazil, and Thailand close behind (Chart 22). J.P. Morgan also announced that it would gradually add Chinese government bonds to its local sovereign bond indexes over a period of 10 months starting February 2020, up to the 10% country cap.15 This move is likely to push the index’s yield lower as Chinese yields are below the current yield on the index. There is some overlap between the drivers of local- and hard-currency sovereign spreads. The most important factor for investors to consider is the direction of emerging market currencies versus the US dollar. This relationship closely tracks inflation differentials between the US and EM economies (Chart 23). Chart 23The Link Between EM Currencies And Inflation Chart 24The USD Is The Most Important Factor To Consider   The top panel in Chart 24 emphasizes this point. It shows that EM local-currency sovereign bonds from a USD perspective have returned -2.8% since the peak in EM currencies in early 2013. This coincides with a time when EM currencies, on a real effective exchange rate basis, weakened against the US dollar (Chart 24, bottom panel). Other drivers of local-currency sovereign yields include commodity prices, global trade, and EM sovereign bond yields. However, this year has witnessed a significant decoupling between local bond yields and these drivers (Chart 25). Chart 25Sustainable Divergence? Chart 26Investors Continue To Hunt For Yield In Emerging Markets Our EM strategists wonder whether we are seeing a “new normal” for EM local bond yields – a paradigm in which they fall, not rise, during periods of slowing global growth and behave similarly to DM yields.16  This, however, would imply that investors view EM local debt as a safe haven rather than a risky asset class. We agree with their conclusion that the recent rally in EM local sovereign bonds – hence the decline in yields – was due, rather, to investors’ hunt for yield in an environment of over $10 trillion of negative-yielding debt (Chart 26). This trend is likely to continue in the short term until there is a sustained pickup in global growth. Once that happens, long-term yields are likely to rise in tandem (Chart 27). Chart 27ALong Term Yields Will Rise When Global Growth Picks Up Chart 27BLong Term Yields Will Rise When Global Growth Picks Up Diversification And Portfolio Impact Investors with a broad mandate can think about EMD as part of their overall portfolios. We analyze how the addition of EMD to a monthly rebalanced “conventional” portfolio, consisting of 50% global equities, 30% global treasurys, and 20% global corporate debt (split equally between investment-grade and high-yield bonds), would have performed since 2003. We found that the incremental additions of each of the three segments of EMD – from 5% to 20% each – produced a higher portfolio risk-adjusted return relative to the conventional portfolio. In all cases, replacing global equities, treasurys, and corporate bonds with EM debt, led either to a higher annualized portfolio return, reduced volatility, or sometimes both (Table 4). Unsurprisingly, given the cyclicality of EM assets, the “enhanced” portfolios have a higher correlation with global equities, as well as with DM corporate bonds (Table 5). Table 4Portfolio Simulation: Risk-Return Profiles (Feb. 2003 – Feb. 2020) Table 5EMD Is Highly Correlated With Global Equities And Corporate Bonds It is important to note, however, that most of the outperformance from the enhanced portfolios – particularly in the most heavily EMD-tilted portfolios – occurred before the slowdown in emerging economies beginning in 2013 (Chart 28). Since 2013, as the USD appreciated against EM currencies, allocating to EM local-currency sovereign bonds detracted from portfolio returns. During this period other EM risk assets, such as equities and corporate bonds, also underperformed their DM counterparts. Chart 28Allocating To EMD Adds Some Value Our Current View Over the past few years, GAA has been structurally negative on EM risk assets – both equity and debt. Productivity levels, far below historical averages, have been a key reason for this view (Chart 29). In a previous Special Report, we argued that productivity needs to mean-revert to its historical average for emerging markets to perform well, but that this is unlikely without structural reform. 17 Chart 29Global Productivity Growth Levels Chart 30Divergence Between Spreads And Growth   Tactically, however, there are times when EM assets can outperform despite the structural headwinds (2016-2017 was an example of this). This could happen again later this year, if global growth continues to rebound. Nonetheless, this optimistic view is on hold due to the risk to global growth in the short term from the coronavirus outbreak. Our global strategists expect global growth to fall to zero in the first quarter of 2020, before picking up throughout the rest of the year – assuming the outbreak is contained within the next few weeks.18  Providing this happens, and our view of global growth reaccelerating pans out, EMD should perform well. Within the asset class, segment selection is key. The environment is likely to be more favorable for EM hard-currency sovereign debt than hard-currency corporate debt or local-currency sovereign bonds. The recent divergence between hard-currency sovereign spreads and growth metric could point to an attractive entry point for investors (Chart 30). We remain cautious on EM corporate bonds, which are vulnerable in the face of sluggish domestic demand in most emerging economies, leading to contracting profits (Chart 31). A weaker USD, when global growth recovers, helped by a dovish stance from the Fed, should keep US financial conditions loose and help EM local-currency sovereign debt perform well (Chart 32). However, relative financial conditions between the US and emerging markets are just as important to monitor. If growth in EM economies fails to pick up, EM currencies could depreciate, putting downward pressure on local-currency sovereign bonds.  Chart 31EM Corporates Face Weak Domestic Demand Chart 32Easier US Financial Conditions Lead To Better EM LC Sovereign Returns We will wait to pull the trigger on this recommendation until we get further clarity regarding the impact on growth of the coronavirus outbreak. Conclusion EMD has grown to become an interesting asset class for allocators, allowing them to capitalize on bonds with higher yields than their DM counterparts. Not only has EMD provided higher returns, it gives equity-like exposure to emerging markets with significantly reduced downside during recessions and market selloffs. We recommend clients view EMD as three separate segments – hard-currency sovereign debt, hard-currency corporate debt, and local-currency sovereign debt – due to the different dynamics that influence each segment. Global growth, the direction of EM currencies versus the US dollar, and EM domestic demand are the three most important overall factors to consider when allocating to any of the segments of EMD.   Amr Hanafy, Senior Analyst amrh@bcaresearch.com   Footnotes 1We use the BIS’s definition of international debt securities (IDS) for hard-currency debt, and domestic debt securities (DDS) for local-currency debt. 2Includes both financial and nonfinancial corporations. 3For the purpose of assessing this segment, we use the broad EM and regional Bloomberg Barclays USD Aggregate Sovereign Indices, which track USD-denominated bonds issued by EM governments. Another commonly used index is the J.P. Morgan Chase & Co.’s EMBI Global Diversified Index, which tracks EM hard-currency sovereign debt, as well as fully owned and explicitly guaranteed quasi-issuers. Additionally, J.P. Morgan Chase & Co’s suite of indices following EM Sovereign debt includes their EMBI+ index. This index is primarily focused on EM sovereign issuers, however with a stricter liquidity requirement for inclusion. The reason why we do not rely on this index is due to its tilt towards LATAM and away from Middle Eastern and Asian economies. 4Please see Emerging Markets Strategy Special Report titled, “A Primer On EM External Debt,” available at ems.bcaresearch.com. 5Commercial index providers treat such distinctions by separating quasi-sovereign entities that are/are not fully owned by governments. For example, J.P. Morgan Chase & Co.’s EMBI Global Diversified Index, probably the most widely used index in tracking EM hard-currency sovereign debt, includes sovereign debt as well as fully owned and explicitly guaranteed quasi-issuers in its index. 6Please see “Fears mount over rise of sovereign-backed corporate debt,” Financial Times, dated January 5, 2016. 7Please see “Dubai World secures deal to restructure $14.6bn debt” Financial Times, dated January 12, 2015. 8For the purpose of assessing this segment, we use the broad EM and regional USD Aggregate Corporate Indices, which track USD-denominated bonds issued by EM corporates. 9Includes Basic Industry, Capital Goods, Communication, Consumer Cyclical, Consumer Non-Cyclical, Energy, Technology, and Transportation sectors. 10Please see Global Asset Allocation Special Report, “High-Yield Bonds: Low Volatility Equities?”, available at gaa.bcaresearch.com 11Annualized returns since 2004. 12Please see Global Asset Allocation Special Report, “Return Assumptions – Refreshed and Refined,” available at gaa.bcaresearch.com 13Please see “Emerging market corporate default and recovery rates, 1995 – 2018,” Moody’s Investors Service, dated January 30, 2019. 14For the purpose of assessing this segment, we use the J.P Morgan GBI-EM global diversified index, an investable benchmark accessible to most investors. This index tracks local-currency bonds issued by EM governments. 15Please see “JP Morgan to add China bonds to GBI-EM indexes from February 2020,” Reuters, dated September 4, 2019. 16Please see Emerging Markets Strategy Weekly Report titled, “EM Local Bonds: A New Normal?”, available at ems.bcaresearch.com. 17Please see Global Asset Allocation Special Report titled, “Return Assumptions – Refreshed and Refined,” available at gaa.bcaresearch.com. 18Please see Global Investment Strategy Report titled, “Markets Too Complacent About The Coronavirus,” available at gis.bcaresearch.com.    

Overweight Global bourses broke down yesterday, succumbing to immense pressure from the coronavirus epidemic gripping South Korea, Italy and Iran. Finally, lower profit growth expectations are weighing on extreme equity market optimism. We remain cautious on the prospects of the broad US equity market and reiterate our January 13 boost to overweight in the global gold mining index versus the MSCI All Country World Index. This portfolio position is already up 16% since the mid-January inception and more gains are in store on the back of the collapse in interest rates and increasing likelihood of fed funds rate cuts. Importantly, in order to gauge the relative attractiveness of this portfolio position at the current juncture, it is instructive to juxtapose global versus US policy uncertainty. Historically, this ratio has been closely correlated with relative share prices and signals that the path of least resistance is higher for gold miners compared with the broad market. Shinning global gold miners are impressive especially given the recent spike in the greenback. As a reminder, the US dollar also flexes its muscles when global uncertainty trumps US uncertainty (bottom panel). Finally, the recent hook up in US economic surprises versus the rest of the world also underpins relative share prices (middle panel). Bottom Line: Stay overweight the global gold mining index via the long GDX:US / short ACWI:US exchange traded funds. From a risk management perspective and in order to protect profits we are setting a trailing stop at the 10% return mark, since inception.