Emerging Markets

Highlights Political and corporate climate activism will increase the cost of developing the resources required to produce and deliver energy going forward – e.g., oil and gas wells; pipelines; copper mines, and refineries. Over the short run, the fastest way for investor-owned companies (IOCs) to address accelerated reductions in CO2 emissions imposed by courts and boards is to walk away from the assets producing them, which could be disruptive over the medium term. Longer term, state-owned companies (SOCs) not facing the constraints of IOCs likely will be required to provide an increasing share of the resources needed to produce and distribute energy. The real difficulty will come in the medium term. Capex for critical metals like copper languishes, just as the call on these metals steadily increases over the next 30 years (Chart of the Week). The evolution to a low-carbon future has not been thought through at the global policy level. A real strategy must address underinvestment in base metals and incentivize the development of technology via a carbon tax – not emissions trading schemes – so firms can innovate to avoid it. We remain long energy and metals exposures.1 Feature And you may ask yourself, "Well … how did I get here?" David Byrne, Once In A Lifetime Energy markets – broadly defined – are radically transforming from week to week. The latest iteration of these markets' evolution is catalyzed by climate activists, who are finding increasing success in court and on corporate boards – sometimes backed by major institutional investors – and forcing oil and gas producers to accelerate CO2 emission-reduction programs.2 Climate activists' arguments are finding increasing purchase because they have merit: Years of stiff-arming investors seeking clarity on the oil and gas producers' decarbonization agendas, coupled with a pronounced failure to provide returns in excess of their cost of capital, have given activists all of the ammo needed to argue their points. Chart of the WeekCall On Metals For Energy Will Increase This activism is not limited to the courts or boardrooms. Voters in democratic societies with contested elections also are seeking redress for failures of their governments to effectively channel mineral wealth back into society on an equitable basis, and to protect their environments and the habitats of indigenous populations. This voter activism is especially apparent in Chile and Peru, where elections and constitutional conventions likely will result in higher taxes and royalties on metals IOCs operating in these states, which will increase production costs and ultimately be passed on to consumers.3 These states account for ~ 40% of world copper output. IOCs Walk Away Earlier this week, Exxon walked away from an early-stage offshore oil development project in Ghana.4 This followed the unfavorable court rulings and boardroom setbacks experienced by Royal Dutch Shell, Chevron and Exxon recently (referenced in fn. 2). While the company had no comment on its abrupt departure, its action shows how IOCs can exercise their option to put a project back to its host government, thus illustrating one of the most readily available alternatives for energy IOCs to meet court- or board-mandated CO2 emissions targets. If these investments qualify as write-offs, the burden will be borne by taxpayers. As climate activism increases, state-owned companies (SOCs) not facing the constraints of IOCs likely will be required to provide an increasing share of the resources – particularly oil and gas – needed to produce and distribute energy going forward. This is not an unalloyed benefit, as the SOCs still face stranded-asset risks, if they invest in longer-lived assets that are obviated by a successful renewables + grid buildout globally. That is a cost that will have to be compensated, when the SOCs work up their capex allocations. Still, if legal and investor activism significantly accelerates IOCs' capex reductions in oil and gas projects, the SOCs – particularly those in OPEC 2.0 – will be able to expand their position as the dominant supplier in the global oil market, and could perhaps increase their influence on price levels and forward-curve dynamics (Chart 2).5 Chart 2OPEC 2.0s Could Expand If Investor Activism Increases Higher Call On Metals At present, there is a lot of talk about the need to invest in renewable electricity generation and the grid structure supporting it, but very little in the way of planning for this transition. Other than repeated assertions of its necessity, little is being said regarding how exactly this strategy will be executed given the magnitude of the supply increase in metals required. Nowhere is this more apparent than in the refined copper market, which has been in a physical deficit – i.e., production minus consumption is negative – for the last 6 years (Chart 3). Physical copper markets in China, which consumes more than 50% of refined output, remain extremely tight, as can be seen in the ongoing weakness of treating charges and refining charges (TC/RC) for the past year (Chart 4). These charges are inversely correlated to prices – when TC/RCs are low, it means there is surplus refining capacity for copper – unrefined metal is scarce, which drives down demand for these services. Chart 3Coppers Physical Deficit Likely Persist Chart 4Chinas Refined Copper Supply Remains TightTheoretically, high prices will incentivize higher levels of production. However, after the last decade’s ill-timed investment in new mine discoveries and expansions, mining companies have become more wary with their investments, and are using earnings to pay dividends and reduce debt. This leads us to believe that mining companies will not invest in new mine discoveries but will use capital expenditure to expand brownfield projects to meet rising demand. In the last decade, as copper demand rose, capex for copper rose from 2010-2012, and fell from 2013-2016 (Chart 5). During this time, the copper ore grade was on a declining trend. This implies that the new copper brought online was being mined from lower-grade ore, due to the expansion of existing projects(Chart 6). Chart 5Copper Capex Growth Remains Weak Chart 6Copper Ore-Quality Declines Persist Through Capex Cycle Capex directed at keeping ore production above consumption will not be sufficient to avoid major depletions of ore supplies beginning in 2024, according to Wood Mackenzie. The consultancy foresees a cumulative deficit of ~ 16mm MT by 2040. Plugging this gap will require $325-$500 billion of investment in the copper mining sector.6 The Case For A Carbon Tax The low-carbon future remains something of a will-o'-the-wisp – seen off in the future but not really developed in the present. Most striking in discussions of the low-carbon transition is the assumption of resource availability – particularly bases metals –in, e.g., the IEA's Net Zero by 2050, A Roadmap for the Global Energy Sector, published last month. In the IEA's document, further investment in hydrocarbons is not required beyond 2025. The copper, aluminum, steel, etc., required to build the generation and supporting grid infrastructure will be available and callable as needed to build all the renewable generation the world requires. The document is agnostic between carbon trading and carbon taxes as a way to price carbon and incentivize the technology that would allow firms and households to avoid a direct cost on carbon. A real strategy must address the fact that most of the world will continue to rely on fossil fuels for decades, as development goals are pursued. Underinvestment in base metals and its implications for the buildout of generation and grids has to be a priority if these assets are to be built. Given the 5-10-year lead times base metals mines require to come online, it is obvious that beyond the middle of this decade, the physical reality of demand exceeding supply will assert itself. A good start would be a global effort to impose and collect carbon taxes uniformly across states.7 This would need to be augmented with a carbon club, which restricts admission and trading privileges  to those states adopting such a scheme. Harmonizing the multiple emissions trading schemes worldwide will be a decades-long effort that is unlikely to succeed. Such schemes also can be gamed by larger players, producing pricing distortions. A hard and fast tax that is enforced in all of the members of such a carbon club would immediately focus attention on the technology required to avoid paying it – mobilizing capital, innovation and entrepreneurial drive to make it a reality. This would support carbon-capture, use and storage technologies as well, thus extending the life of existing energy resources as the next generation of metals-based resources is built out. In addition, a carbon tax raises revenue for governments, which can be used for a variety of public policies, including reducing other taxes to reduce the overall burden of taxation. Lastly, a tax eliminates the potential for short-term price volatility in the pricing of carbon – as long as households and firms know what confronts them they can plan around it.  Tax revenues also can be used to reduce the regressive nature of such levies. Investment Implications The lack of a coherent policy framework that addresses the very real constraints on the transition to a low-carbon economy makes the likelihood of a volatile, years-long evolution foreordained. We believe this will create numerous investment opportunities as underinvestment in hydrocarbons and base metals production predisposes oil, natural gas and base metals prices to move higher in the face of strong and rising demand. We remain long commodity index exposure – the S&P GSCI and GSCI Commodity Dynamic Roll Strategy ETF (COMT), which is optimized to take advantage of the most backwardated commodity forward curves in the index. These positions were up 5.3% and 7.2% since inception on December 7, 2017 and March 12, 2021, respectively, at Tuesday's close. We also remain long the MSCI Global Metals & Mining Producers ETF (PICK), which is up 33.9% since it was put on December 10, 2020. Expecting continued volatility in metals – copper in particular – we will look for opportunities to re-establish positions in COMEX/CME Copper after being stopped out with gains. A trailing stop was elected on our long Dec21 copper position established September 10, 2020, which was closed out with a 48.2% gain on May 21, 2021. Our long calendar 2022 vs short calendar 2023 COMEX copper backwardation trade established April 22, 2021, was closed out on May 20, 2021, leaving us with a return of 305%.   Robert P. Ryan Chief Commodity & Energy Strategist rryan@bcaresearch.com Ashwin Shyam Research Associate Commodity & Energy Strategy ashwin.shyam@bcaresearch.com   Commodities Round-Up Energy: Bullish OPEC 2.0 offered no surprises to markets this week, as it remained committed to returning just over 2mm b/d of production to the market over the May-July period, 70% of which comes from the Kingdom of Saudi Arabia (KSA), according to Platts. While Iran's return to the market is not a given in OPEC 2.0's geometry, we have given better than even odds it will return to the market beginning in 3Q21 and restore most of the 1.4mm b/d not being produced at present to the market over the course of the following year. OPEC itself expects demand to increase 6mm b/d this year, somewhat above our expectation of 5.3mm b/d. Stronger demand could raise Brent prices above our average $63/bbl forecast for this year (Chart 7). Brent was trading above $71/bbl as we went to press. Base Metals: Bullish BHP declared operations at its Escondida and Spence mines were running at normal rates despite a strike by some 200 operations specialists. BHP is employing so-called substitute workers to conduct operation, according to reuters.com, which also reported separate unions at both mines are considering strike actions in the near future. Precious Metals: Bullish The Fed’s reluctance to increase nominal interest rates despite indications of higher inflation will reduce real rates, which will support higher gold prices (Chart 8). We agree with our colleagues at BCA Research's US Bond Strategy that the Fed is waiting for the US labor market to reach levels consistent with its assessment of maximum employment before it makes its initial rate hike in this interest-rate cycle. Subsequent rate changes, however, will be based on realized inflation and inflation expectations. In our opinion, the Fed is following this ultra-accommodative monetary policy approach to break the US liquidity trap, brought about by a rise in precautionary savings due to the pandemic. In addition, we continue to expect USD weakness, which also will support gold and precious metals prices. We remain long gold, expecting prices to clear $2,000/oz this year. Ags/Softs: Neutral Corn prices fell more than 2% Wednesday, following the release of USDA estimates showing 95% of the corn crop was planted by 31 May 2021, well over the 87% five-year average. This was in line with expectations. However, the Department's assessment that 76% of the crop was in good-to-excellent condition exceeded market expectations. Chart 7 Chart 8 Footnotes 1     Please see Trade Tables below. 2     Please see OPEC, Russia seen gaining more power with Shell Dutch ruling and EXCLUSIVE BlackRock backs 3 dissidents to shake up Exxon board -sources published by reuters.com June 1, 2021 and May 25, 2021. 3    Please see Chile's govt in shock loss as voters pick independents to draft constitution published by reuters.com May 17, 2021, and Peru’s elite in panic at prospect of hard-left victory in presidential election published by ft.com June 1, 2021.  Peru has seen significant capital flight on the back of these fears.  See also Results from Chile’s May 2021 elections published by IHS Markit May 21, 2021 re a higher likelihood of tax increases for the mining sector.  The risk of nationalization is de minimis, according to IHS. 4    Please see Exxon walks away from stake in deepwater Ghana block published by worldoil.com June 1, 2021. 5    Please see OPEC 2.0's Production Strategy In Focus, which we published on May 20, 2021, for a recap our how we model OPEC 2.0's strategy.  It is available at ces.bcaresearch.com. 6    Please see Will a lack of supply growth come back to bite the copper industry?, published by Wood Mackenzie on March 23, 2021. 7     Please see The Challenges and Prospects for Carbon Pricing in Europe published by the Oxford Institute for Energy Studies last month for a discussion of carbon taxes vs. emissions trading schemes.     Investment Views and Themes Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Trades Closed in 2021 Summary of Closed Trades

Special Report

Highlights Asset Management Regulation (AMR) represents a critical and successful structural reform that is defusing risks in the most hazardous parts of China’s credit system. This bodes well for long-term sustainability of the nation’s financial system and, hence, its long-term economic outlook. That said, the sheer size of risky products and shadow banking makes it impossible to reduce systemic risk without hampering overall credit origination. AMR will dampen bank and shadow banking credit growth further and the credit impulse will be negative by year-end. As a result, China's growth will decelerate. The risk-reward profile of Chinese stocks remains poor. Favor Chinese local currency government bonds as yields will drop further. Feature Chart 1China’s Growth Is Set To Decelerate China’s broad credit and money growth have relapsed substantially. Given that they have historically been reliable leading indicators of business cycles (Chart 1), the question is: how far will credit growth decelerate. When gauging the magnitude of a money/credit slowdown, one should not only look at borrowing costs but also at the willingness and capacity of creditors to extend credit. In this context, it is essential to examine the impact of Asset Management Regulation (AMR) in China on both bank and non-bank credit growth. Please refer to Box 1 below for a more detailed discussion on AMR.     BOX 1 What Is AMR? AMR (Asset Management Regulation) was introduced in 2018 to mitigate financial system risks, increase transparency of financial products, and, hence, enhance investor protection. Financial institutions (banks and non-banks) were originally obliged to meet AMR requirements by the end of 2020. However, after the pandemic broke out, this term was extended to the end of 2021. The main objectives of AMR are: To restrict financial institutions from dodging financial regulations and prevent them from engaging in regulatory arbitrage. To prohibit financial institutions from providing other financial organizations with “channels” for evading regulatory requirements. To preclude banks from investing in high-risk assets. To forbid financial institutions from providing explicit or implicit guarantees for the principal and return on asset management products. AMR non-compliant products need to be either terminated or revamped to become AMR compliant before December 31, 2021. Assessing the value of outstanding AMR non-compliant products will help to gauge the actual impact of AMR on credit growth over the course of this year. A portion of banks’ wealth management products (WMP) and single fund trust products are AMR non-compliant and will need to be terminated or revamped. Commercial banks’ WMPs represent fund investment and management plans developed, designed and sold by commercial banks to individuals or institutions. In China, individual investors are the main customers for banks’ WMPs. In 2020, individual investors accounted for more than 99% in number of investors and 87% in investment amounts.1 The outstanding amount of WMPs is presently RMB 25 trillion. Single fund trusts have one investor – usually a bank or another financial institution. Given the disclosure regulation for single fund trusts is much looser than other fund trusts, it was prevalently used by financial institutions, including banks, to channel funds into investments to achieve regulatory arbitrage. Chart 2China Has Not Yet Deleveraged AMR represents regulatory tightening and will negatively affect bank and non-bank credit growth over the course of this year. In this report we examine what its impact will be on broad credit growth as banks and shadow banking attempt to comply with AMR by end of December this year. Authorities in China have been conducting well-thought-out surgical reforms – AMR being the cornerstone of these – to curb and restructure the risky elements of the credit system. By doing so, they have already dramatically reduced systemic risk in the financial system. Regardless of how deft and precise these reforms have been, they will continue to weigh on bank and shadow banking credit growth. The basis is that the sheer size of risky products and shadow banking makes it impossible to reduce systemic risk without hampering overall credit origination. It should also be noted that China has not yet deleveraged (Chart 2). How Large Are AMR Non-Compliant Assets? We reckon that AMR’s effect on broad credit is mainly through its impact on commercial banks’ Wealth Management Products (WMP) and single fund trusts. S&P Global2 estimates that by the end of 2020, banks will still have RMB 8.5 trillion in off-balance sheet WMP to restructure.  Single fund trusts’ assets stood at RMB 7.7 trillion in March 2021. However, to avoid double counting, flows from banks to trust funds (“bank-trust cooperation”) should be deducted from this value. The basis is that channeling funds by banks via trust companies is already captured in banks’ WMP statistics. Overall, non-compliant AMR assets that need to be revamped by year-end are as follows: Banks’ non-compliant WPM          8.5 trillion Single fund trust assets excluding “bank-trust cooperation”                   1.2 trillion Total                                          RMB 9.7 trillion This RMB 9.7 trillion represents 3.6% of total social financing (TSF) excluding equity issuance and 4.2% of private credit. The latter is defined as TSF excluding equity and central and local government bond issuance as well as special bonds.  Chart 3China: Various Borrowing Costs SP Global2 estimates that around RMB 5 trillion WMP will be revamped and made AMR compliant during this year. To put this figure into perspective, banks revamped RMB 4.8 trillion in 2020 and RMB 5.7 trillion in 2019. This will leave RMB 3.5 trillion of non-compliant WMP that banks are likely to take on their balance sheet before year-end. Even in the case of revamped WMP and single fund trusts, there will be unintended consequences for borrowers. In particular, the cost of borrowing could rise and/or the maturity of loans could be shortened. Both will weigh down on economic activity in general, and investment in the real economy in particular.   With full transparency and no implicit guarantee from banks, investors will require higher interest rates to invest in these products (Chart 3). In addition, investors will opt for shorter maturities of these products. Impact On Bank Credit… Chart 4China: Bank Loan Approvals And Bank Credit Impulse As banks take these AMR non-compliant WMP onto their balance sheets, their assets will automatically expand even though they will not originate new loans/provide financing to the real economy. The estimated RMB 3.5 trillion of WMP is equivalent to 1.5% of commercial bank broad credit and 1.2% of their assets. Hence, AMR will reinforce the deceleration in new credit origination. Both bank assets and broad bank credit will slow and their impulses will contract further (Chart 4).   Importantly, bringing these assets onto their balance sheet will require banks to both (1) allocate more capital to support these new assets and (2) increase provisions for the portion of these assets that are non-performing. The non-performing share of these AMR-non-compliant assets could be significant given that funds from off-balance sheet WMP were often invested in high-risk, high-return projects. These often represent claims on risky businesses, including property developers and local government financing vehicles (LGFV). In brief, there were reasons why banks did not initially put these assets on their balance sheets and doing so now will not be inconsequential. Overall, this move will hinder commercial banks’ ability and willingness to originate new credit, i.e., to provide new funding to the real economy (Chart 4). …And Shadow Banking Chart 5 demonstrates that shadow banking credit – comprised of trust loans, entrust loans, and unrealized banker acceptance bills – has been contracting. Outstanding shadow banking credit at RMB 23.9 trillion makes up 9% of TSF excluding equity issuance. Single fund trust loans – please refer to Box 1 above for more information – are the most vulnerable part of shadow banking to AMR. Despite their having contracted since 2017, single fund trust assets excluding “bank-trust cooperation” still amount to RMB 1.2 trillion or 0.5% of TSF, excluding equity issuance (Chart 6). Chart 5China’s Shadow Banking Continues To Shrink Chart 6Single Fund Trusts Are The Most Vulnerable To AMR Regulation     This type of financing will continue to shrink, weighing on aggregate credit flow. Although investors in these products might reinvest their funds in AMR-compliant funds, they will demand higher interest rates to offset higher credit risk. The basis is that full transparency will inform them that the trust companies and banks can neither guarantee principal nor interest on their investments. Higher interest rates demanded by investors in trust funds or their reduced financing will affect borrowers that rely on funding from this source. Specifically, trust funds investment in property developers and LGFV has been and will continue to shrink (Chart 7).      Impact On Property Developers And LGFV Property developers and LGFV are among the most vulnerable segments to reduced financing because of AMR. Trust companies have meaningful exposure to both real estate developers and LGFV. RMB 2.3 trillion in trust funds are invested in real estate and RMB 1.2 trillion in government projects, mostly representing claims on LGFV. Trust companies’ claims to both segments have been and will continue contracting (Chart 7). Property developers and LGFV are not only vulnerable to curtailed funding due to AMR but also from authorities’ campaign to limit their debt. Three Red Lines policy for property developers imposes caps on their debt. In addition, bank regulators have imposed limits on banks’ claims on property developers as well as residential mortgages (Chart 8, top panel). Loans are capped at 40% for the largest state-owned lenders, while banks’ mortgage lending should be no more than 32.5% of large banks’ outstanding credit. The regulations are even more rigorous for smaller banks. For smaller banks, caps on loans to real estate and mortgage loans are 27.5% and 20%, respectively.3 Banks’ credit to property developers and household mortgages are growing at a historically low pace and will likely decelerate further (Chart 8, bottom panel). To sum up, banks and shadow banking will curtail their exposure to property developers and LGFV. Consequently, these credit-intensive sectors will have to shrink their capital spending and construction activity. The latter will have ramifications for raw materials and industrial sectors exposed to traditional infrastructure and construction. Chart 7Trust Funds’ Exposure To Property Developers And LGFVs Chart 8Banks’ Exposure To Property Developers And Residential Mortgages   Investment Conclusions On the positive side, AMR represents critical and successful structural reform that is defusing risks in the most hazardous parts of China’s credit system. This bodes well for long-term sustainability of the nation’s financial system and, hence, its long-term economic outlook. Nevertheless, this regulatory tightening along with clampdown on the property market and local government debt will weigh on the Chinese business cycle over the next six-to-nine months: Private credit growth will continue downshifting and its impulse will turn negative, weighing on credit-exposed sectors (Chart 9). Although the private credit impulse is unlikely to reach -10% of GDP like it did in 2018, it will likely turn negative by year-end. Our guess it might be negative 3-4 % of GDP later this year. Chart 9China: Private Credit Impulse Will Turn Negative By Year-End Chart 10China: Fiscal Spending Impulse Will Be Modestly Positive In 2021   Public sector credit – measured as borrowing by central and local government, including special-purpose bonds – will continue decelerating according to bond quotas for this year. Still, higher government revenue will offset the slump in government borrowing so that government spending will grow in 2021 from a year ago. In aggregate, the fiscal spending impulse for all of 2021 will be positive at 1.6% of GDP (Chart 10). Overall, the fiscal spending impulse of 1.6% of GDP in 2021 will not offset the private credit impulse that we reckon to be about negative 3-4% of GDP. The upshot will be a modestly negative aggregate credit and fiscal spending impulse. The latter will be slightly worse than the readings of this indicator during the 2011 and 2014-15 slowdowns but more positive than in 2018 (please refer to Chart 1 above). This heralds a non-trivial business cycle slowdown. The latter will be concentrated in areas that usually benefit from credit and fiscal stimulus. Construction activity and traditional infrastructure spending are the most vulnerable areas. This entails that Chinese demand for raw materials will disappoint and base metals prices are vulnerable. With regard to investment strategy, investors should continue favoring Chinese local currency government bonds over stocks. As the economy decelerates, bond yields will drift lower. Share prices remain vulnerable. Chart 11 illustrates that net EPS revisions for the MSCI China A-share index has rolled over but has not yet dropped to their previous lows. Our hunch that EPS slowdown is not yet fully priced into the Chinese onshore equity market. Concerning MSCI China Investable non-TMT stocks, they have rolled over at their previous high (Chart 12). Given the negative corporate profit outlook, the risk-reward is unattractive both in absolute terms and relative to global equities. Chart 11Chinese Stocks: EPS Growth Expectations Will Downshift Further Chart 12An Intermediate-Term Top In Chinese Non-TMT Stocks?   In the long run, however, the de-risking of the credit system is bullish for Chinese share prices. Declining systemic financial risks entail a lower equity risk premium. Consequently, equity valuations will ultimately be re-rated. Arthur Budaghyan Chief Emerging Markets Strategist arthurb@bcaresearch.com Qingyun Xu Associate Editor Qingyun@bcaresearch.com   Footnotes 1 2020 Bank’s Wealth Management Product Report 2 Source: SP Global "China Banks May Still Have RMB3 Trillion In Shadow Assets By Year-End Deadline." 3 https://www.cbirc.gov.cn/cn/view/pages/ItemDetail.html?docId=955074&ite…   Cyclical Investment Stance Equity Sector Recommendations

Special Report

Dear client, This week, I am conducting a BCA Academy Marcroeconomic seminar in the Middle East. In lieu of our regular report, we are publishing a piece written by my colleague Jeremie Peloso. In it, Jeremie explores how to adjust valuation metrics to build country and sector selection tools which can be deployed to manage global equity portfolios. I trust you will find that this report provides a useful approach to equity selection. Best Regards, Mathieu Savary Chief European Strategist   Highlights We introduce our Combined Mechanical Valuation Indicator for European equities to identify extreme valuations at the country and sector level. At the country level, the historical track record of relative valuations as an alpha-generating tool is mixed; however, they demonstrate impressive predictive power at the sector level on a 3- to 12-month time horizon. A trading strategy consisting of a basket of the five cheapest relative valuations generates excess returns with high batting averages. The current reading from our Combined Mechanical Valuation Indicator suggests investors should overweight the following European sectors: consumer discretionary relative to both Swedish and British counterparts, tech relative to Australian counterparts, communications relative to Spanish counterparts, and utilities relative to Italian counterparts. Also, favor UK energy stocks relative to their Eurozone competitors. Feature European equities have been underperforming their foreign peers for the past 10 years (Chart 1). The persistently lower profitability of European stocks partly explains their subpar performance; a DuPont decomposition of RoE reveals how Europe’s economic malaise affects corporate profitability (Chart 2). Chart 1Structural Underperformance From The Past... Chart 2... And The Future The Eurozone’s excessively large capital stock is chief among these culprits (Chart 2, bottom panel). It suggests that a large proportion of the capital stock in the Eurozone is misallocated which, in turn, hurts profit margins and renders the Euro Area’s asset turnover inferior to that of other countries. Compared to the US, greater economic rigidities and lower market power and concentration in Europe also hurt profitability. On net, these forces indicate that the case for overweighting European equities on a structural investment horizon (5 to 10 years) remains weak. Despite the poor long-term outlook, European stocks could still perform well on both a tactical and cyclical investment horizon. We currently recommend a modest overweight in European stocks for cyclical investors. One of our main investment themes for the remainder of 2021 is that European growth will surprise to the upside, once the re-opening of economic activity in the Eurozone gets fully underway, supported by the rapid recent progress of vaccination campaigns. This process will cause a re-rating of European assets. Our recent work shows that positive changes in economic surprises translate into generous returns for European equities and EUR/USD. Moreover, prolonged accommodative monetary policies via low rates and the ECB’s PEPP program, as well as continued fiscal support via the NGEU recovery fund, will be supportive for European assets in absolute terms. However, there are risks to our upbeat view, which we explored last week. They are as follows: (1) a slowdown in the Chinese economy, (2) a global credit impulse deterioration, and (3) inflation surges that are faster than expected. While none of these risks constitute our base case scenario, they could derail the positive cyclical environment we anticipate for European equities. In order to diversify portfolio risk away from traditional cyclical factors, this Special Report presents a mechanical valuation framework for European equities to identify high-probability attractive excess returns on a 3- to 12-month time horizon. At the country level, the historical track record of relative valuation as a selection tool is mixed; however, it demonstrates impressive predictive power at the sector level. Therefore, this method provides an attractive starting point for sector selection. The Mechanics Of The Mechanical Approach The starting point of this analysis is to select different valuation metrics. We opt for the following measures commonly accepted by the investment community: Price-to-earnings, Forward price-to-earnings, Price-to-sales, Price-to-book,  Price-to-cash flows, Long-term growth in earnings. Next, we detrend each valuation measure by subtracting its 5-year moving average. We subsequently compute the difference between the detrended valuation metrics of the Euro Area MSCI equity benchmark and its chosen counterpart. For example, the calculation for the price-to-earnings ratio (P/E) with the US is as follows: Valuation Gap = (Euro Area P/E - 5-year m.a.) - (US P/E - 5-year m.a.) Then, we divide each of the valuation gaps shown above by their 5-year moving standard deviation: Mechanical Indicator = Valuation Gap / (5-year moving standard deviation of VG) The resulting valuation indicator mean-reverts and oscillates between +/- 2 standard deviations (Chart 3). We repeat this process for each valuation metric across 15 countries (including the All Country World and emerging markets MSCI indices) and the 10 GICS sectors. Considering the importance of relative sectoral biases, we create two versions of the mechanical indicators for the purpose of country analysis: a regular market-cap weighted version and a sector-neutral one, in which we weight all 10 GICS sectors equally. As Chart 4 illustrates, the differences in sector composition between the Eurozone and other regions lead to a sector-neutral valuation metric that deviates substantially from its market-cap weighted counterpart. Importantly, the sector-neutral mechanical indicators perform better on average than the market-cap weighted versions, thus reinforcing the importance of relative sectoral biases when it comes to equity valuation. Chart 3Mechanical Valuation Indicator Example Chart 4Sector Composition Matters Finally, given the sheer amount of computations performed, we only present the summary output from our analysis. The appendix, which starts on page 11, displays the detailed results for each of the valuation metrics, countries, and sectors. A Well-Oiled Mechanical Tool? Simple valuation measures make unreliable market timing tools. However, they are useful at extreme levels, which is precisely how the mechanical indicator is supposed to be used. The next step of our analysis is to assess our methodology and see where it displays predictive power. For this purpose, we back-tested trading rules relying on outlying readings of the relative Mechanical Valuation Indicator. More specifically, we calculated the common currency (US$) excess returns over 3-, 6-, and 12-month horizons generated by the following: Going long (overweight) European stocks, when they stood at 1 and 1.5 standard deviations on the cheap side of fair value. Going short (underweight) European stocks, when they stood at 1 and 1.5 standard deviations on the expensive side of fair value. We define excess returns as the returns in excess of the average returns observed over the past 10-year period. In other words, we want to ensure that the mechanical approach delivers more alpha than a passive buy-and-hold strategy. We use the 1.5 standard deviation threshold rather than the 2-sigma hurdle because of the lack of sufficient observations at the 2-standard deviation bar. If we had stuck to the 2-sigma threshold, the results from the back-test would not have been reliable, despite a sample with history going back to 2003. Table 1 presents the indicator’s batting average at the country level for all the valuation metrics - that is, the number of times both trading rules generated positive excess returns as a percent of the total number of signals. Table 1Mechanical Valuation Indicator (Sector-Neutral) Historical Track Record: Country Level The results are mixed. Individually, none of the metrics display batting averages that significantly exceed 50% and none of the valuation metrics seem to perform uniformly across either time horizons or trading rules. On the bright side, we observe an improvement in excess returns between the +/- 1 and 1.5 standard deviation signals, especially when the mechanical indicators signal that European equities are the most expensive. Looking more closely at each valuation metric reveals that the long-term expected growth in earnings and the price-to-cash flows provided much better signals than the forward P/E and the price-to-book metrics. We repeat the same exercise at the sector level by calculating mechanical indicators for European sectors relative to comparable sectors from other regions - for example, European industrials relative to US or Chinese industrials. The results displayed in Table 2 consist of the average excess returns and batting averages across all sectors. The results for each sector can be found on page 19.    Table 2Mechanical Valuation Indicator Historical Track Record: Sector Level The historical track record of valuation-based trading rules yields much better results for sector selection than for country picking. All of the valuation metrics provide respectable predictive ability except for the long-term expected growth in earnings. In fact, the indicator generates positive excess returns more than two-thirds of the time; in half of the cases when the indicator fails to generate alpha, the Mechanical Valuation Indicator is computed using the long-term expected growth in earnings. Furthermore, the batting averages are above the 50% mark often, except over 12-month time horizons. Strength In Numbers: Combining The Signals The mixed results obtained from applying trading rules based on our mechanical indicator at the country level suggest we could improve the predictive power of this framework. Since individual valuation metrics do not cut it, we combine them into a simple average. Table 3Combined Mechanical Valuation Indicator (Sector-Neutral) Historical Track Record: Country Level At the country level, the results are once again disappointing. As can be seen from Table 3, the quality of the signals from our combined mechanical indicator is not consistent across the board. The predictive power of the combined signals only appears to be effective when European equities are 1-sigma cheap or 1.5-sigma expensive. When the combined mechanical indicator is 1.5 standard deviations away from fair value on the expensive side, which, admittedly, is not a very common occurrence, going short (underweight) European equities deliver excess returns of 4.2%, 3.2%, and 2.6% over  3-, 6- and 12-month time horizons, respectively. Table 4Combined Mechanical Valuation Indicator Historical Track Record: Sector Level Despite this disappointment, the mechanical indicator once again truly shines at the sector level. Combining the valuation metrics, excluding the long-term expected growth rate of earnings (which, as we showed does a poor job), provides an excellent predictive power on all fronts (Table 4). All the excess returns are positive, and the batting averages are satisfying, especially on the 3-month and 6-month time horizons. The most impressive performance came from the mechanical indicator signaling European equity sectors were 1.5-sigma cheap. Out of 61 occurrences, following the signal resulted in earned excess returns of 3.3% and 4.8% on average over a 6- and 12-month time horizon, respectively. Importantly, the batting averages were both close to 60%. Bottom Line: Our Combined Mechanical Valuation Indicator is a useful tool, especially for sector selection in a global portfolio. It sports an impressive historical track record and allows us to identify pockets of attractive relative valuation that generate alpha for investors on a 3- to 12-month time horizon. Investment Implication What is the current message from our Combined Mechanical Valuation Indicator? Chart 5Combined Mechanical Valuation Indicators (Sector-Neutral): Country Level At present, the approach only sends two signals at the +/- one-sigma threshold at the country level and both stand on the cheap side of fair value (Chart 5). According to the sector-neutral mechanical indicator, the European MSCI equity benchmark is cheap compared to emerging markets and Chinese benchmarks. And, while not at extremes, US and global equities are still expensive relative to Eurozone stocks. Chart 6 provides the current reading from the mechanical indicator for each sector. Chart 6ACombined Mechanical Valuation Indicators: Sector Level Chart 6BCombined Mechanical Valuation Indicators: Sector Level Chart 7Favor UK Energy Stocks Vs. European Ones A few things stand out. First, there appears to be no extreme relative valuations within materials. Second, European energy stocks turn out to be expensive relative to all other regions included in the analysis, especially against energy stocks out of China and the UK. In fact, it makes a compelling case for investors to underweight Euro Area energy stocks relative to UK counterparts (Chart 7). Third, within the communications sector, Eurozone stocks are cheap against all their counterparts except for German ones. The relative valuation does not, however, stand at an extreme. Finally, if we were to select the five strongest signals, we would select the following pairs: Overweight European consumer discretionary stocks relative to Swedish counterparts Overweight European communications stocks relative to Spanish counterparts Overweight European tech stocks relative to Australian counterparts Overweight European consumer discretionary stocks relative to UK counterparts Overweight European utilities stocks relative to Italian counterparts This basket should deliver positive excess returns over a 3- to 12-month time horizon (Chart 8). Chart 8Going With The Strongest CMVI Signals   Jeremie Peloso, Associate Editor JeremieP@bcaresearch.com   Appendix A The tables below present the historical track record of the sector-neutral mechanical valuation indicator for each of the valuation metrics at the country level. Euro Area vs. USEuro Area vs. All Country World Euro Area vs. Emerging Markets Euro Area vs. Germany Euro Area vs. France Euro Area vs. Italy Euro Area vs. Spain Euro Area vs. The Netherlands Euro Area vs. UK Euro Area vs. Sweden Euro Area vs. Switzerland Euro Area vs. Japan Euro Area vs. Canada Euro Area vs. Australia Euro Area vs. China   Appendix B The tables below present the historical track record of the mechanical valuation indicator for each of the valuation metrics at the sector level. Industrials Materials Consumer Discretionary Consumer Staples Energy Financials Technology Communications Utilities Health Care Appendix C The tables below present the historical track record of the sector-neutral combined mechanical valuation indicator (CMVI) at the country level.   Euro Area vs. US Euro Area vs. All Country World Euro Area vs. Emerging Markets Euro Area vs. Germany Euro Area vs. France Euro Area vs. Italy Euro Area vs. Spain Euro Area vs. The Netherlands Euro Area vs. UK Euro Area vs. Sweden Euro Area vs. Switzerland Euro Area vs. Japan Euro Area vs. Canada Euro Area vs. Australia Euro Area vs. China     Appendix D The tables below present the historical track record of the Combined Mechanical Valuation Indicator (CMVI) at the sector level. Industrials Materials Consumer Discretionary Consumer Staples Energy Financials Technology Communications Utilities Health Care   Footnotes

Highlights Our long-term FX REER models suggest the dollar remains overvalued, especially against the Chinese yuan.  The cheapest currencies are the yen and the Russian ruble. The Scandinavian currencies are surprisingly expensive, according to these models. This has been due to falling relative productivity. Other notable expensive currencies are the Hong Kong dollar and Saudi riyal. That said, we do not expect the peg in the former to break anytime soon. Our limit-sell on the yen was triggered at 109. Place stops at 112. We are looking to buy a basket of petrocurrencies that include the COP and RUB. These have significantly lagged the rise in oil prices.  Feature This week’s report focuses on our long-term fair value models. But a few words first on currency developments. In our view, currency markets are likely to remain driven by five important trends in the coming months. A rotation of growth from the US to other parts of the world (dollar bearish): This has been the dominant theme that has played out since the peak in the DXY index in March. The manufacturing sector in other countries first caught up to the buoyancy we saw in the US, and their service sectors are now recovering as the world vaccinates its population and reopens. In the developed world, Japan, which has been a laggard, could witness a bout of positive surprises. Market focus on inflation, and the potential of an overshoot (dollar bearish): Most market participants have been paying close attention to the inflation overshoot in the US, and whether it is transitory. Currency markets however, specifically the dollar, have been paying close attention to the inflation differential between the US and other countries, and what that means for relative real rates. A rising inflation differential between the US and its trading partners has been negative for the dollar (Chart I-1). We have noted that the US will continue to provide relative upside surprises in inflation as the US output gap closes ahead of other countries. This has been in part due to the most generous fiscal stimulus in the developed world. Chart I-1The Dollar And Relative Inflation Move Opposite Ways A Federal Reserve that stays ultra-accommodative (dollar bearish): Most market participants are again focused on the Fed tapering and what that will mean for asset markets. The reality is that the Fed has started to lag many other central banks, like the Bank of Canada, the Reserve Bank of New Zealand and the Bank of England in tapering asset purchases. This could suggest it would also lag in the speed and magnitude of lifting policy rates in the medium term. This will keep US real rates depressed relative to many of its trading partners. A risk event (dollar bullish): We have been highlighting that a risk event, like a market reset, is a strong positive for the dollar, given the negative correlation with risk assets (Chart I-2). A dollar that remains expensive (dollar bearish): Our medium-term (12-18 month) target for the DXY index is 80. This will bring the currency towards fair value, according to our purchasing power parity models. As we highlighted last week, the trade balance in the US continues to deteriorate, which is one of the symptoms of an overvalued currency. Chart I-2The Dollar And Risk Assets Move Opposite Ways Despite our bearish dollar view, it is important not to overstay our welcome. This week, we are updating our long-term models, another technical tool we use to help us navigate FX markets. These models are mostly driven by relative productivity, but we have also fine-tuned the models for each currency to account for other factors such as terms-of-trade shocks, real rate differentials and proxies for global risk aversion. These models cover 22 currencies, incorporating both G10 and emerging FX markets. The dollar remains expensive according to these models (Chart I-3). Chart I-3The US Dollar Remains Expensive It is important to note that these models are very poor timing tools and are not designed to generate short- or medium-term forecasts. Instead, they reflect imbalances in the current equilibrium fair value of a currency. For example, a currency might be flagged as overvalued now, but a productivity boom in the next few years could allow the currency fair value to gravitate higher. So will a commodity boom. From a technical perspective, these models are like the ones we published in our last report, but with a very important change – the weights assigned in calculating relative productivity are based on dynamic trade weights. This has allowed China (which has much better productivity growth) to impact the currency fair values significantly. For all countries, the variables are highly statistically significant and are of the right signs. Finally, as housekeeping, we were triggered into a short USD/JPY position this week as our limit-sell at 109 was touched. The yen is one of the cheapest currencies according to these models. It will also benefit from all of the five key drivers for currency markets we listed above, especially real rates that are likely to stay very favorable in Japan, compared to the US (Chart I-4). Chart I-4Less Inflationary Pressures In Any Japanese Economic Rebound The US Dollar Chart I-5 The dollar is expensive by 7% according to the long-term fair value model. This is despite the 13% drop in the US dollar DXY index since the March 2020 highs. In hindsight, strong reversals in the dollar occur when the currency is about two-standard deviations above the mean, which occurred with last year’s rally. Our bias is that the dollar has entered a multi-year downtrend, which will only be supercharged by expensive valuations. The big driver for the uptrend that started in 2011 was positive real interest rate differentials. As US real rates continue to rollover, relative to its G10 counterparts, this will lower the greenback’s fair value. The Euro Chart I-6 The euro is slightly cheap according to our fundamental models. More importantly, the euro’s fair value has been rising in recent quarters. This has been driven by a nascent improvement in the trade balance (and current account balance), following the Covid-19 crisis. Historically, when the euro has hit its fair value bands, it has tended to mean revert. Therefore, this model does a better job of catching intermediate turns in the euro, compared to the US dollar model. Our bias is that the long-term fair value for the euro sits near 1.35, something that should continue to be reflected in future model updates. The Yen Chart I-7 The fair value of the yen has been relatively flat over the last few years. Given that the real exchange rate has not fluctuated much either, the yen has been chronically undervalued by about one standard deviation below the mean. The yen is cheap by most measures of relative prices. We believe the yen sits at a beautiful juncture. A pickup in economic activity will keep the fair value rising, from an improvement in the current account. Meanwhile, any deterioration in economic data will lead to higher risk aversion and a higher fair value (the yen is a risk-off currency). We are short USD/JPY as of 109 this week.   The British Pound Chart I-8 This model shows that the pound is fairly valued, while cable remains cheap by most of our other models. That said, at fair value, the pound can still overshoot to at least 1.5 standard deviation above/below the mean, as it has in prior episodes. The key reason the pound is not cheap in this model is due to a deterioration in the UK’s productivity growth, relative to its trading partners. In this iteration of the model, China’s larger share of British trade has exacerbated the downtrend in the fair value. However, a turnaround seems underway, as the UK puts the Brexit woes behind it (and Scottish independence is not an immediate concern). The Canadian Dollar Chart I-9 The loonie has overshot its fair value. More importantly, the fair value for the Canadian dollar has been falling since the peak of the commodity cycle in 2011. If we are indeed entering a new commodity super-cycle, then the model should begin to turn around, and assign a higher fair value to the loonie. However, Canada’s terms of trade will face strong headwinds as we move away from fossil fuels, especially oil. As such, the productivity gains in other sectors (such as metals) that will benefit from new green investments will need to be sufficiently high to offset falling productivity in crude oil.  The Australian Dollar Chart I-10 The Australian dollar has been rising along with the improvement in its fair value. The rising fair value has been due to the exceptional rise in commodity prices (iron ore and coal) that have boosted the current account. However, like the Canadian dollar, the fair value of the Aussie has also been dropping in recent years on the back of previously depressed commodity prices. Given the growing importance of liquified natural gas in Australia’s export mix, we believe terms of trade will remain a tailwind for the Aussie over the longer term. The New Zealand Dollar Chart I-11 The kiwi is slightly more expensive than its antipodean neighbor. But like other commodity currencies, its fair value has fallen in recent years. The catalyst has been the drop in commodity prices and the fall in relative real rates. More recently, the fair value of the kiwi has taken a positive turn as real rates improve, and risk aversion recedes. With the RBNZ striking a hawkish tone, this remains positive for the kiwi for now, but could adversely impact financial conditions later. The Swiss Franc Chart I-12 On a fundamental basis, the Swiss franc is as cheap as the yen, with our models showing it as about one standard deviation undervalued. The biggest driver for the rise in the fair value of the franc has been the structural trade surplus, driven by rising productivity. The Swiss franc is traditionally a defensive currency. As such, the fair value has taken a small hit due to the fall in the gold-to-oil ratio, a proxy for risk aversion. Should the market experience some turbulence in the coming months, the franc will benefit. The Swedish Krona Chart I-13 The Swedish krona is showing up as expensive in our models, together with the Norwegian krone. Paradoxically, on a PPP basis, the Swedish krona is one of the cheapest currencies in our universe. The key model inputs for the Swedish krona are interest rate differentials and relative productivity trends. With the rise of China as a trading partner, the productivity differential for Sweden has fallen even more steeply in this iteration of the model. It also means that the currency is no longer massively undershooting fair value, as had been the case in previous iterations. The Norwegian Krone Chart I-14 Like the Swedish krona, the Norwegian krone is showing up as expensive in our models. However, it is one of the cheapest currencies on a PPP basis, which presents a paradox. We will be looking at the Norwegian economy in-depth next week, to help explain this paradox. A more immediate explanation is that the trade balance for Norway has been nosediving in recent years, which helps explain why the model judges the currency as becoming incrementally expensive. With the rise of China as a trading partner, the productivity differential for Norway has also fallen. The Chinese Yuan Chart I-15 The Chinese yuan is currently at about one standard deviation below fair value. Since the history of our model, the fair value of the yuan has been mostly rising. This is driven by rising relative productivity in China. Concurrently, real interest rates in China have also shot up, which has led to a strong rally in the Chinese RMB. We expect the RMB to keep appreciating in the coming years, as the fair value keeps rising.  The Brazilian Real Chart I-16 The Brazilian real is slightly above fair value, according to our fundamental models. Meanwhile, the fair value has been falling since 2011, in line with other commodity currencies. However, if we are indeed in a new commodity super cycle, the fair value of the real should start to rise. The Mexican Peso Chart I-17 The Mexican Peso is trading a nudge above fair value, but the fair value has been rising since 2019. This means going forward, we could see a rising peso, coinciding with a rapid rise in its fair value. The peso is highly cyclical, so two key drivers have been working in favor of the currency. First, a falling in risk aversion (proxied by a decline in the gold-to-oil ratio) has been positive. Meanwhile, the cumulative current account will also continue to improve should global growth remain strong in the near term, especially US growth. The Chilean Peso Chart I-18 The Chilean peso is currently at fair value, according to our fair value model. The fair value of the Chilean peso has been falling in recent years, but this decline was even sharper when Chinese productivity gains were given a greater weight in the modelling exercise. Going forward, Chilean exports of copper will be in a structural uptrend, due to the green technology revolution. As such, the fair value of the peso should begin to gradually rise. The Colombian Peso Chart I-19 The Colombian peso is cheap, and so constitutes an attractive play if oil prices remain strong in the medium term. The reason is that it has one of the strongest correlations to oil prices among commodity currencies. That said, structurally, the fair value of the Colombian peso has been falling, like many other petrocurrencies. The South African Rand Chart I-20 The South African rand is now trading slightly below its fair value, a positive contrast to the BRL which is slightly expensive. Meanwhile, the fair value of the rand is gradually picking up, after a structural decline over the last decade.   The correlation between precious metal prices and the South African rand is picking up again, as the current account moves back into surplus. We are positive gold (and silver) as inflation hedges. Meanwhile, platinum and palladium will continue to benefit from a push towards better environmental standards among traditional autos. The Russian Ruble Chart I-21 The Russian ruble is now sitting around one standard deviation below its fair value. We are constructive on oil, which will boost the fair value of petrocurrencies, including the Russian ruble. Meanwhile, real interest rates are at relatively high levels in Russia, even though this does not have significant explanatory power. Given cheap valuations, we are looking to buy a petrocurrency basket, including the RUB and the COP. The Korean Won Chart I-22 The Korean won has underperformed this year, and has been trading at a negligible divergence from its fair value over the last few years. The fair value of the Korean won has also been flat over the years. This suggests that Korean productivity growth has kept pace with its trading partners. Going forward, it also suggests the next move in the Korean won is likely to be driven by the trend in the currencies of its trading partners, especially China and the US. The Philippine Peso Chart I-23 The Philippine peso is expensive by about one standard deviation. This has been partly due to a decline in the fair value of the peso, a process that began in 2015.  The Philippine peso is one of the few currencies whose REER tends to have well-defined and long cycles that last 5-8 years. It will be important to watch if the recent appreciation in the currency has more room to run, given expensive valuation. The Singapore Dollar Chart I-24 The Singaporean dollar is another currency whose REER tends to have long cycles, probably a feature of the managed float. The Singaporean dollar is a defensive currency, and so the appreciation in other emerging market currencies has brought relative valuations back towards fair value. The Hong Kong Dollar Chart I-25 The REER of HKD has been rising in recent years, meaning inflation in Hong Kong SAR has been outpacing that elsewhere. This has made the HKD expensive, according to our models. However, the fair value has started to fall suggesting productivity gains in the city state have also been lagging, probably a result of political unrest. That said, we expect the peg to remain in place for some time, as we highlighted in a previous special report. The Saudi Riyal Chart I-26 The fair value of the Saudi Riyal has been falling for quite a while on declining relative productivity. This has made the Riyal incrementally expensive. However, oil prices are currently elevated, which means it might take much more stretched valuations to begin to cause greater tensions for the peg.   Chester Ntonifor Foreign Exchange Strategist chestern@bcaresearch.com Trades & Forecasts Forecast Summary Core Portfolio Tactical Trades Limit Orders Closed Trades