Emerging Markets

Special Report

Highlights This report presents our framework for estimating Chinese capital outflows on a monthly basis, which investors can use as a real-time indicator to monitor the risk of another serious episode of capital flight. We also provide a monthly estimate of illicit capital outflow, which we find is negatively correlated with “on balance sheet” capital flows. This implies that Chinese residents alternate their use of the two channels in their attempt to move money out of the country. Our monitoring framework suggests that outflow pressure is more likely to ease than intensify if a trade deal is struck over the coming few weeks or months, especially given the rise in CNY-USD since early-November. However, we have identified a low-odds but high-impact scenario in which a shaky trade deal with the U.S. generates an unstable equilibrium that could ultimately escalate into a major Chinese capital outflow event. This could prove to be a highly destabilizing event for investors, and thus bears close monitoring. Feature Fears of a new round of capital outflow from China re-emerged in the second half of 2018 as USD-CNY approached 7, a psychologically important level for many investors (Chart 1). The last episode of significant capital outflows from China occurred in late-2015 following the PBOC’s devaluation of the RMB, and the sharp spike in volatility that resulted had a contagious effect for global financial markets. Chart 1A Near Miss Late Last Year In the very near term, the risk of a similar event appears to be low given the material trade talk-driven decline in USD-CNY that has occurred over the past five months. However, several news reports over the past year concerning the possible risk of another episode of capital flight underscore that China’s cross-border capital flow statistics are misunderstood by financial market participants. This raises the risk that investors either fail to anticipate a capital outflow event in the future or exaggerate the odds of one occurring. In this report we present our framework for estimating Chinese capital outflows on a monthly basis, which investors can use as a real-time indicator to monitor the risk of another serious episode of capital flight. We also adjust the typical measure of short-term capital flow derived from the quarterly balance of payments to account for cross-border RMB settlement, and present an estimate of illicit capital outflow that suggests Chinese residents alternate their use of legal and illegal channels in their attempt to move money out of the country. We then combine these three direct measures of capital flow with two indicators of expected RMB depreciation to further augment our monitoring efforts. We conclude by noting that while outflow pressure is more likely to ease than intensify if a trade deal is struck over the coming few weeks or months, we have identified a low-odds but high-impact scenario in which a shaky trade deal with the U.S. generates an unstable equilibrium that could ultimately escalate into a major Chinese capital outflow event. This scenario is not part of our base case outlook, but could prove to be a highly destabilizing event for investors and thus bears close monitoring. Defining Short-Term Capital Flow From The Balance Of Payments Table 1 presents China’s balance of payments (BOP) for the four quarters ending in Q3 2018, with all items shown on a net basis. The table is organized in a way that provides a helpful refresher on the formulation of the balance of payments, namely that the current account (“CA”, made up of the trade balance and primary & secondary income) plus the sum of the capital account (“KA”), the financial account (“FA”), and a balancing item (referred to as net errors & omissions, “NEO”) is equal to 0, when capital and financial outflows are recorded with a minus sign. Current account surpluses necessarily involve net financial outflows (i.e., investment); whereas current account deficits must be funded by financial inflows (i.e., borrowing). Table 1 highlights that what financial market participants typically refer to as “capital” flows are actually recorded in the financial account of the balance of payments. While derivatives are included in the table for the sake of completion, in practice they are usually quite small (as is the case for the actual “capital” account). Table 1China’s Balance Of Payments The bottom panel of Table 1 indicates that the balance of payments formula can be rearranged so that it represents how many market participants tend to define total and short-term capital outflows from a balance of payments perspective. As we will show in the next section of the report, this re-arrangement of the balance of payments formula is an essential element in building a more frequent proxy of short-term capital flow. We define short-term capital flow from the balance of payments as the combination of portfolio investment, other investment, and net errors & omissions. The bottom panel shows that by adding reserve assets (“RA”) to the current account (“CA”), the right hand side of the BOP equation becomes the sum of direct investment (“DI”), portfolio investment (“PI”), other investment “OI”, and net errors & omissions (“NEO”). Since direct investment tends not to be driven by short-term economic behavior and is normally not influenced by foreign exchange expectations or fluctuations, the formula can be further arranged to isolate short-term capital outflows on the right-hand side: Current Account + Changes in Reserve Assets + Direct Investment ≈ (Portfolio Investment + Other Investment + Net Errors & Omissions)*-1 Or using our line item notation, CA + RA + DI ≈ -PI - OI - NEO The formula above is expressed as an approximation rather than an identity simply because it excludes the capital account (“KA”) and financial derivatives (“FD”). As can be seen in Table 1, the net value of adding the four quarter rolling total of CA + RA + DI to PI + OI + NEO is US$ 3.3 billion; adding KA + FD (-0.35 and -2.95 billion US$, respectively) would result in a value of 0. Chart 2 shows this relationship visually; and highlights that both series are nearly identical. Chart 2Short-Term Capital Flow As Defined By The BOP Building A Better Proxy Of Short-Term Capital Flow The balance of payments approach is a useful starting point for measuring short-term capital flow, but it has two important drawbacks: Timeliness: Balance of payments data are reported in quarterly frequency, and often with a lag. This is inadequate for most investors, particularly when market participants are concerned that a crisis or crisis-like conditions may be emerging. This is the primary disadvantage of the BOP approach. Failure to account for cross-border RMB settlement: The balance of payments approach implicitly assumes that a current account surplus in China will automatically result in the importation of foreign exchange, but this assumption is no longer fully valid. Cross-border RMB settlement now accounts for part of China’s foreign trade settlement, reaching more than 30% during the 2015/2016 period. Compared with its peak level, RMB settlement as a share of total foreign trade has fallen over the past two years, but still accounts for 19% today (Chart 3). To more precisely gauge China’s capital outflows, cross-border RMB settlement should be removed from the current account surplus, because trade payments settled in RMB would not involve the receipt of foreign currency. This offsetting current account discrepancy would still show up in the balance of payments under net errors & omissions, but that would have the effect of distorting our definition of short-term capital flow. Chart 3Analysts Need To Adjust The Current Account For Cross-Border RMB Settlement Chart 4 illustrates the difference between our quarterly definition of short-term capital flow and the series adjusted for cross-border RMB settlement. The chart shows that the two series are quite similar for most of the past decade, with the notable exception of the 2015/2016 period. The adjusted series suggests that the intensity of China’s episode of capital flight did not peak in 2015, but rather late in 2016. This is consistent with domestic commentary at the time,1 and implies that the PBOC faced headwinds in their attempt to stem capital outflows that were even worse than has been generally acknowledged. Chart 4After Adjusting For Cross-Border Settlement, Outflow Intensity Only Peaked In Late-2016 Unfortunately for investors, dealing with the lack of timeliness in the release of China’s balance of payments statistics is a more challenging endeavor. This problem cannot be resolved with simple adjustments to the quarterly data, and instead requires the building of a proxy for short-term capital flow based on the BOP equation but using monthly statistics. Investors can proxy our adjusted quarterly balance of payments-based measure of short-term capital flow on a monthly basis. As we referenced above, the key to constructing a monthly capital flow estimate lies with the re-arrangement of the balance of payments equation such that short-term capital flow is expressed as being approximately equal to the sum of the current account, direct investment, and the change in reserve assets (when outflows of the latter two series are recorded as negative values). Table 2 highlights that high quality monthly series are available to act as proxies for these three balance of payments components, after accounting for cross-border RMB settlement and the following two additional adjustments: Table 2Components Of BCA’s Monthly China Capital Outflow Indicator Services Balance: The trade balance accounts for the vast majority of the current account of most countries, and this is also true in the case of China. An underappreciated fact about China’s trade balance is that it has shrunk considerably over the past several years, due to what is now a sizeable services deficit. Some market commentators who are aware of the services deficit point to it as evidence that China’s net importation of services is laying the groundwork for its “new economy” (via eventual import substitution), but the reality is that travel (i.e. net tourism spending) accounts for over 80% of it (Chart 5). For the purposes of our monthly capital flow proxy, a sizeable services deficit is a complication that must be accounted for, given that China’s monthly trade statistics (and most monthly trade data globally) represent the trade in goods, not the trade in services. Since most of the fluctuations in the trade balance occur due to net trade in goods, we include the history of the quarterly services balance in our monthly indicator as a structural variable, and extend the most recent quarterly value into the current quarter as a simplifying assumption. Currency Valuation Effect on Official Reserves: Foreign exchange reserves in the balance of payments are calculated by the historical cost method, whereas the highly followed monthly official foreign exchange reserve data released by the PBOC is measured using market value. Changes in its balance, in addition to genuine changes in foreign exchange reserve assets, also reflect revaluation effects caused by fluctuations in the foreign exchange market. To dampen these effects, we include foreign exchange reserves in our monthly capital flow proxy in SDR terms rather than in U.S. dollars, rebased to the value of the underlying U.S. dollar series as of December 2018. Chart 5Travel (i.e. Tourism) Accounts For The Majority Of China's Services Deficit Chart 6 presents our quarterly balance of payments-based capital flow measure (adjusted for cross-border capital flow) with our monthly proxy, based on the series shown in Table 2 and the adjustments noted above. Divergences between the series exist in level terms, but panel 2 shows that our monthly proxy does a good job capturing the trend in the quarterly series. The only major exception to this occurred at the beginning of 2016, when our monthly proxy fell sharply relative to the adjusted quarterly BOP version. Chart 6Our Monthly Proxy Captures The Trend In Quarterly Capital Flows This sharp decline is a bit of a mystery; it can be traced to the official reserves series, and either suggests that capital outflow was materially worse in Q4 2015 and Q1 2016 than officially recognized, or that China suffered outsized losses from the risky asset portion of its reserve portfolio during that period. However, the first explanation is at odds with the evidence noted earlier that the intensity of capital flight seems to have peaked in late-2016, and the second explanation is inconsistent with the history of financial market returns over the past decade. We noted in a February 2018 Special Report that risky U.S. assets (almost entirely stocks) accounted for as much as 9.5% of China’s foreign reserve assets in the summer of 2015,2 and it is true that U.S. equity returns were quite negative from December 2015 to February 2016. But this was certainly not the first and only period of extreme U.S. equity market volatility to occur since 2010, raising the question of why this sharp decline in official reserves only occurred in 2015/2016. Future research on the topic of Chinese capital flows will aim to reconcile the difference between our monthly proxy and our adjusted quarterly balance of payments series during this period, but for now we are confident that the former contributes meaningfully to our understanding of the latter, particularly on a rate of change basis. Import Over-Invoicing: A Third Measure Of Short-Term Capital Outflow Investors need to track both legal and illicit capital flows. Our first two measures of short-term capital flow were based on an attempt to track the legally allowable movement of funds out of China. However, illicit capital outflow is an acknowledged problem in China, which tends to occur through the practice of import over-invoicing.3 Chart 7 presents our estimate of import over-invoicing for China, based on a methodology articulated by Global Financial Integrity, a U.S. non-profit organization that provides analysis of illicit financial flows globally (see Appendix A). The chart highlights two important points: Chart 7Illicit Capital Outflows: Another Way That Money Leaves China Illicit outflows have increased significantly over the past 2 years following China’s capital control crackdown, particularly in Q3 2018 following the announcement of the second round of U.S. import tariffs against China. Panel 2 of Chart 7 illustrates that there is a negative correlation between “on balance sheet” capital flows and illicit capital outflows, implying that Chinese residents alternate their use of the two channels in their attempt to move money out of the country. This underscores the importance of monitoring both channels on an ongoing basis. Investment Conclusions Table 3 brings together the three measures of short-term capital flow that we have laid out above, as well as two indicators of expected RMB depreciation (Chart 8): net settlement of foreign exchange by Chinese banks (see Appendix B), and the 3-month moving average of the percent deviation of CNH-USD (offshore RMB) from CNY-USD (onshore RMB). Altogether, the series shown in Table 3 form the basis of our capital outflow monitoring efforts, and we plan on updating these series regularly to gauge whether outflow pressure is increasing. Table 3Dashboard For Monitoring Short-Term Capital Flows Chart 8Two Indicators Capturing Expectations Of Severe RMB Depreciation For now, only our measure of illicit capital outflow is flashing a warning sign, and the timing of the recent spike in the measure appears to be closely connected with the trade war with the U.S. This implies that outflow pressure is more likely to ease if a trade deal is struck over the coming few weeks, as we expect will occur. However, we noted in a March 6 joint Special Report with our Geopolitical Strategy service that a deal with only slight concessions from China may stand on shaky ground and that tariff rollbacks will be limited or non-existent.4 This would ensure elevated policy uncertainty in the aftermath of the agreement and would raise the probability of a relapse into another trade war ahead of the 2020 U.S. election. In this scenario we would be watching the indicators shown in Table 3 closely for signs that increasing pessimism about the long-term state of sino-U.S. relations is causing the capital outflow “dam” built by policymakers following the 2015/2016 episode to buckle. Our monitoring framework suggests that the odds of a major capital flight event are currently low. But a shaky trade deal with the U.S. could change that. It is not part of our base case outlook, but onshore concerns of a renewed trade war with the U.S. next year could theoretically become self-fulfilling, if another major episode of capital flight were to weaken the RMB in a way that could even remotely be construed as a violation of the yuan stability pact that will reportedly be part of any agreement between the U.S. and China. While this would in no way entail a purposeful devaluation by Chinese policymakers to boost trade competitiveness, it could nonetheless provide an excellent excuse for President Trump to reinstate damaging economic pressure on China in the midst of what is likely to be a highly competitive re-election campaign. This could, in turn, produce a feedback effect that magnifies the original desire to move capital out of China, and would likely prove to be a highly destabilizing event for global financial markets. Stay tuned!   Qingyun Xu, CFA, Senior Analyst qingyunx@bcaresearch.com Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com     Appendix A Measuring Import Over-Invoicing In this report we use one of the two methodologies employed by Global Financial Integrity to measure import over-invoicing in China, which compares a country’s reported trade statistics with that of its global trade partners.5 Using the IMF’s Direction of Trade Statistics data, we deflate Chinese import data measured on a C.I.F. (cost insurance and freight) basis to an F.O.B. (free on board) basis using an assumed freight and insurance factor of 10%. Then, we use Hong Kong re-export data to adjust global exports to China for re-exported trade through Hong Kong. The formula is listed below: Chinese Import Over-invoicing = [(Chinese Imports From The World)/1.1] - Adjusted Global Exports To China   Appendix B The Onshore Market For Foreign Exchange A poorly understood fact about China’s capital/financial account regime is that a material amount of foreign exchange reserves are now held by enterprises and individuals. Most investors are familiar with China’s old foreign exchange settlement policy (established formally in 1993), which prohibited enterprises from retaining foreign currency. Exporters receiving foreign currency as payment for goods and services had to sell all foreign exchange receipts to designed banks, and purchase foreign exchange from these banks when needed to make payments to offshore suppliers. Thus, while this policy was in effect, the PBOC held all China’s foreign exchange reserves and official reserves equaled total reserves. However, since the early-2000s, this policy has been gradually withdrawn. Since its complete abolishment in 2012, foreign exchange retained by enterprises and residents has increased materially. Chart B1 shows the impact of these changes on the bank foreign exchange settlement and sale rates. The settlement rate represents enterprises’ sale of foreign exchange to banks as a share of their total foreign exchange receipts in a given month, while the sale rate represents banks’ sale of foreign exchange to enterprises as a share of enterprises’ total foreign exchange payments. The chart shows that the settlement rate has dramatically dropped since 2012 (from 70% to less than 50%). We can also see there were spikes in the settlement rate and sale rate in August 2015 (in contrary directions) when the PBOC devalued the RMB, implying that the demand for forex and presumably the expectation of further RMB depreciation was severe. Chart B1The Evolution Of China’s Domestic Foreign Exchange Market​​​​​​​   Given this, we view net FX settlement (enterprises’ sale of foreign exchange to banks minus banks’ sale of foreign exchange to enterprises) as a reasonable proxy of expected RMB depreciation, and have included it as part of our capital flow monitoring framework.         1 “China’s capital outflow is still intensifying”, Reuters China Finance and Economics Column, December 19, 2016. 2 Please see China Investment Strategy Special Report, “Demystifying China’s Foreign Assets”, dated February 28, 2018, available at cis.bcaresearch.com. 3 Import over-invoicing occurs when an importer (in country A) attempts to evade capital controls by colluding with an exporting entity (in country B) to falsify the reported value of goods imported into country A from country B. The importer “overpays” for the goods in question and, usually through an intermediary, moves the surplus funds into the importer’s offshore account. Please see https://www.gfintegrity.org/issue/trade-misinvoicing/ for more information about the mechanics of and motivations behind trade misinvoicing. 4 Please see Geopolitical Strategy and China Investment Strategy Special Report, “China-U.S. Trade: A Structural Deal?”, dated March 6, 2019, available at cis.bcaresearch.com. 5 “Illicit Financial Flows to and from 148 Developing Countries: 2006-2015”, Global Financial Integrity, January 2019. Cyclical Investment Stance Equity Sector Recommendations

Highlights Global equities will remain rangebound for the next month or so, but should move decisively higher as economic green shoots emerge in the spring. A revival in global growth will cause the recent rally in the U.S. dollar to stall out and reverse direction, setting the stage for a period of dollar weakness that could last until the second half of next year. Rising inflation will force the Fed to turn considerably more hawkish in late-2020 or early-2021. This will cause the dollar to surge once more. The combination of a stronger dollar and higher interest rates will trigger a recession in the U.S. in 2021, which will spread to the rest of the world. Investors should maintain a bullish stance towards global equities for the next 12 months, but look to reduce exposure at some point next year. Feature Stocks Temporarily Stuck In The Choppy Trading Range We argued at the end of February that global equities and other risk assets would likely enter a choppy trading range in March as investors nervously awaited the economic data to improve.1 Recent market action has been consistent with this thesis, with the MSCI All-Country World Index falling nearly 3% at the start of the month, only to recoup its losses over the past few days. We expect stocks to remain in a holding pattern over the coming weeks, as investors look for more evidence that global growth is bottoming out. The U.S. dollar is a countercyclical currency, meaning that it moves in the opposite direction of global growth (Chart 1). This countercyclicality stems from the fact that the U.S. economy is more geared towards services than manufacturing compared with the rest of the world. This makes the U.S. a low-beta play on global growth (Chart 2). As such, when global growth accelerates, capital tends to flow from the U.S. to the rest of the world, translating into more demand for foreign currency and less demand for dollars. Chart 1The Dollar Is A Countercyclical Currency Chart 2The U.S. Is A Low-Beta Play On Global Growth Given the dollar’s countercyclical nature, it is not surprising that the slowdown in global growth over the past 12 months has given the greenback a lift. The broad trade-weighted dollar has strengthened by almost 8% since February 2018, putting it near the top of its post 2015-range (Chart 3). Chart 3The Dollar Has Gotten A Lift From Global Growth Disappointments Stocks Will Rally And The Dollar Will Weaken Starting In The Spring We expect the U.S. dollar to strengthen over the coming weeks as global economic data continues to underwhelm. However, an improvement in leading economic indicators in the spring will set the stage for a reacceleration in global growth and a decline in the dollar in the second half of this year. The combination of stronger growth and a weaker dollar later this year should be highly supportive of global equities. Equity investors with a 12-month horizon should overlook any near-term weakness and maintain a bullish bias towards stocks. We do not have a strong view on U.S. versus international equities at the moment, but expect to upgrade the latter once we see more confirmatory evidence that global growth is bottoming out. Equity investors with a 12-month horizon should overlook any near-term weakness and maintain a bullish bias towards stocks.   A Stronger China Will Lead To A Weaker Dollar Our expectation that the dollar will weaken in the second half of this year hinges on what happens to China. The deceleration in global growth in 2018 was largely the consequence of China’s deleveraging campaign. China’s slowdown led to a falloff in capital spending throughout the world. Weaker Chinese growth also put downward pressure on the yuan, pulling other EM currencies lower with it (Chart 4). All this occurred alongside an escalation in trade tensions, further dampening business sentiment. Chart 4EM Currencies Are Off Their Early 2018 Highs While it is too early to signal the all-clear on the trade front, the news of late has been encouraging. A recent Bloomberg story described how Trump watched approvingly as Asian stocks rose and U.S. futures rallied following his decision to delay the scheduled increase in tariffs on Chinese goods.2 As a self-professed master negotiator, Trump needs to secure a deal with China before next year‘s presidential election, while also convincing American voters that the deal was concluded on favorable terms for the United States. Reaching a deal with China early on in his term would have been risky if the agreement had failed to bring down the bilateral trade deficit — an entirely likely outcome given how pro-cyclical U.S. fiscal policy currently is.  At this point, however, Trump can crow about making a great deal with China while reassuring voters that the product of his brilliance will be realized after he has been re-elected. This means that we are entering a window over the next 12 months where Trump will want to strike a deal. For their part, the Chinese want as much negotiating leverage with the Trump administration as they can muster. This means being able to convincingly demonstrate that their economy is strong enough to handle the repercussions from turning down a trade deal that fails to serve their interests. Since the credit cycle is the dominant driver of Chinese growth, this requires putting the deleveraging campaign on the backburner. Admittedly, credit growth surprised on the downside in February. However, this followed January’s strong showing. Averaging out the two months, credit growth appears to be stabilizing on a year-over-year basis. Conceptually, it is the change in credit growth that correlates with GDP growth.3 Thus, merely going from last year’s pattern of falling credit growth to stable credit growth would still imply a positive credit impulse and hence, an uptick in GDP growth. In practice, we suspect that the Chinese authorities will prefer that credit growth not only stabilize but increase modestly. In the past, this outcome has transpired whenever credit growth has fallen towards nominal GDP growth (Chart 5). The prospect of a rebound in credit growth in March was hinted at by the PBOC, which spun the weak February data as being caused by “seasonal factors.” Chart 5Historically, China Has Scaled Back On Deleveraging When Credit Growth Has Fallen Close To Nominal GDP Growth Europe: Down But Not Out Stronger growth in China will help European exporters. Euro area domestic demand will also benefit from a rebound in German automobile production, the winding down of the “yellow vest” protests in France, and incrementally easier fiscal policy. In addition, the ECB’s new TLTRO facility should support credit formation, particularly in Italy where the banks remain heavily reliant on ECB funding. Our expectation that the dollar will weaken in the second half of this year hinges on what happens to China. Euro area financial conditions have eased significantly over the past three months, which bodes well for growth in the remainder of the year. It is encouraging that the composite euro area PMI has rebounded to a three-month high. The expectations component of the euro area confidence index has also moved up relative to the current situation component, which suggests further upside for the PMI in the coming months (Chart 6). Chart 6Easing Financial Conditions Bode Well For Euro Area Growth The selloff in EUR/USD since last March has been largely driven by a decline in euro area interest rate expectations (Chart 7). If euro area growth accelerates in the back half of the year, the market will probably price back in a few rate hikes in 2020 and beyond. Chart 7EUR/USD Sell-Off Has Been Driven By Falling European Rate Expectations What Will The Fed Do? Of course, the degree to which a steeper Eonia curve benefits EUR/USD will depend on what the Fed does. The 24-month discounter has fallen from over +100 bps in March 2018 to -25 bps today, implying that investors now believe that U.S. short rates will fall over the next two years (Chart 8). Chart 8The Fed's Dovish Messaging Has Worked... Almost Too Well We expect the Fed to raise rates more than what is currently priced into the curve, thus justifying a short duration position in fixed-income portfolios. However, the Fed’s newfound “baby step” philosophy will probably translate into only two hikes over the next 12 months. Such a gradual pace of Fed rate hikes is unlikely to prevent the euro from appreciating against the dollar starting in the middle of this year, especially in the context of a resurgent global economy. We do not expect any major inflationary pressures to emerge in the near term. In contrast to the euro, the yen should depreciate against the dollar in the back half of this year. The yen is a “risk-off” currency and thus tends to weaken whenever global risk assets rally (Chart 9). The government is also about to raise the sales tax again in October, a completely unnecessary step that will only hurt domestic demand and force the Bank of Japan to prolong its yield curve control regime. We would go long EUR/JPY on any break below 123. Chart 9The Yen Is A Risk-Off Currency A Blow-Off Rally In The Dollar Starting In Late-2020 What could really light a fire under the dollar is if the Fed began raising rates aggressively while the global economy was slowing down. In what twisted parallel universe could that happen? The answer is this one, provided that inflation rose to a level that evoked panic at the Fed. We do not expect any major inflationary pressures to emerge in the near term. The growth in unit labor costs leads core inflation by about 12 months (Chart 10). Thanks to a cyclical pickup in productivity growth, unit labor cost inflation has been trending lower since mid-2018. However, as we enter late-2020, if the labor market has tightened further by then, wage growth will likely pull well ahead of productivity growth, causing inflation to accelerate. Chart 10Decelerating Unit Labor Costs Will Dampen Inflationary Pressures For The Time Being All things equal, higher inflation is bearish for a currency because it implies a loss in purchasing power relative to other monies. However, if higher inflation spurs a central bank to hike policy rates by more than inflation has risen – thus implying an increase in real rates – the currency will tend to strengthen. Chart 11 shows the “rational expectations” response of a currency to a scenario where inflation suddenly and unexpectedly rises by one percent relative to partner countries and stays at this higher level for five years while nominal rates rise by two percent. The currency initially appreciates by 5%, but then falls by 2% every year, eventually finishing down 5% from where it started.4 The yen should depreciate against the dollar in the back half of this year. The real world is much messier of course, but we suspect that the dollar will stage a final blow-off rally late next year or in early-2021 (Chart 12). Since the Fed will be hiking rates in a stagflationary environment at that time, global growth will weaken, further boosting the dollar. The resulting tightening in both U.S. and global financial conditions will likely trigger a global recession and a bear market in stocks. Investors should maintain a bullish stance towards global equities for the next 12 months, but look to reduce exposure at some point next year.   Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com Footnotes 1      Please see Global Investment Strategy Weekly Report, “Gretzky’s Doctrine,” dated March 1, 2019. 2      Jennifer Jacobs and Saleha Mohsin, “Trump Pushes China Trade Deal to Boost Markets as 2020 Heats Up,” Bloomberg, March 6, 2019. 3      Recall that GDP is a flow variable (how much production takes place every period), whereas credit is a stock variable (how much debt there is outstanding). By definition, a flow is a change in a stock. Thus, credit growth affects GDP and the change in credit growth affects GDP growth. 4      The 2% annual decline in the currency is necessary for the real interest parity condition to be satisfied. Strategy & Market Trends MacroQuant Model And Current Subjective Scores Tactical Trades Strategic Recommendations Closed Trades

We continue to expect copper prices to increase in the near term, as China’s credit cycle bottoms and DM central banks soften their monetary-policy stance. Fiscal and monetary stimulus in China also will be supportive of base metals prices going forward. The evolution of the Sino - U.S. trade negotiations remains a risk to our view, given how important the outcome of these talks will be for investors’ expectations and sentiment. Markets appear to be discounting a positive outcome. Anything that scuppers these talks – or results in a no-deal outcome – will be a negative for base metals, copper in particular. Our tactical long copper position is up by 1.2% since we initiated it last week. Highlights Energy: Overweight. Russian oil companies are expected to keep production lower until July, when the current OPEC 2.0 production-cutting agreement now in place expires. We expect the deal will be extended to year-end.1 Separately, the risk of a complete shutdown in Venezuela’s oil industry rose significantly, as a power failure in most of the country all but eliminated potable water supplies and significantly reduced oil exports. Base Metals/Bulks: Neutral. High-grade iron-ore prices got a boost this week as Vale was ordered to temporarily suspend exports from its primary port at Guaiba Island terminal in Rio de Janeiro state, according to Metal Bulletin’s Fastmarkets.2 The price-reporting agency’s 62% Fe Iron Ore Index rose $1.46/MT at $85.25/MT Tuesday. Precious Metals: Neutral. Spot gold is back above $1,300/oz, on the back of monetary policy easing among important central banks. This also is supporting base metals globally (see below). Ags/Softs: Underweight. Grain markets continue to drift sideways, awaiting definitive news re Sino - U.S. trade talks, specifically when presidents Xi and Trump will meet to finalize a deal (see below). Separately, wheat and corn inventories are expected to rise on the back of higher supplies and lower exports, the USDA forecast in its latest world supply-demand estimates. Feature Recent data releases confirm our view that global growth will remain weak in 1Q19 and early 2Q19. This will continue to put downward pressure on cyclical commodities – chiefly base metals and oil (Chart of the Week). Chart of the WeekGlobal Growth Slows In 1Q19 The persistence of the slowdown provoked major central banks to adopt a dovish stance in the short-term. This is easily seen in the recent actions by the U.S. Fed, the European Central Bank (ECB), the Bank of Canada (BoC) and the Reserve Bank of Australia (RBA), all of which have communicated a pause in their rate normalization policies.3 At the moment, the frail global growth is partly balanced by expectations of a positive outcome re the ongoing Sino - U.S. trade negotiations (Chart 2). In the coming months, we expect the effect of accommodative DM monetary policy combined with an expansion in China’s credit (more on this below) and fiscal stimulus – i.e., tax cuts announced earlier this month amounting to almost $300 billion (~ 2 trillion RMB) meant to support policymakers’ GDP growth targets – will go a long way toward reversing the earlier contraction. The effect of these policy decisions will be apparent in 2H19. Chart 2China Growth To Hook Higher China’s Credit Cycle Bottomed In December 2018 The evolution of China’s credit cycle remains a central pillar to our view commodity demand growth in 2H19 will surpass consensus expectations. The massive growth reported in China’s January credit statistics revived investors’ expectations that China’s banks will re-open the credit valves as they did in 2016.4 In our view, this number does signal a bottom in China’s credit cycle, and implies Chinese – and indirectly EM – growth will bottom sometime this year. However, we still are not expecting a complete blowout credit expansion this year. We continue to believe Chinese policymakers will focus on stabilizing credit in 1H19 with moderate increases in supply, and start increasing stimulus in 2H19 or 2020 in order to maximize its effect later in 2020 ahead of the 100th anniversary of the founding of the Chinese Communist Party (CCP) in 2021. The soft February credit number released this week supports this argument.5 China’s Credit Cycle Matters For Base Metals Demand The relationship between China’s credit cycles and base metal prices endures and remains robust. We measure China’s aggregate credit using bank and non-bank claims on non-financial enterprises, households, local and central governments, and non-bank financial institutions. This corresponds to adding outstanding central and local government bonds to China’s Total Social Financing (TSF).6 The annual change in aggregate credit – or its impulses – do not perfectly capture the cycles in global base metal demand. These variables provide interesting signals about the direction and magnitude of movements in credit, however, they do not track base metals’ price cycles accurately and consistently (Chart 3). Chart 3Metals Price Cycles Don't Track Changed In China's Credit To decompose this variable into its trend and cycle, we use a proxy of the credit cycle constructed using the Hodrick-Prescott and Hamilton filters, and the standardized 12-month credit impulse (Chart 4).7 Chart 4China's Credit Cycle Proxy We find that our credit cycle proxy Granger causes base metal prices, import volume and industrial activity (Table 1).8 On average, it leads these variables by 4-6 months (Chart 5). Hence, we believe our credit cycle proxy provides valuable information about future commodity demand in China. Table 1China Credit Cycle Correlations In fact, when regressing copper prices and the LMEX against it, we found that 60% and 58% of the variation in copper prices and the LMEX, can be explained by the linear relationship with our China credit cycle proxy, respectively (Chart 6). Chart 6China's Credit Cycle and Metals Prices Given the leading property of China’s credit cycles with respect to industrial activity and metal prices, we included this new proxy in our Global Industrial Activity (GIA) index.9 This improves the correlation of our index with copper prices (Chart 7). Chart 7Credit Cycle Improves BCA's GIA Currently, our models suggest copper prices should increase in the coming months as China’s credit cycle bottoms and DM central banks soften their monetary policy stance. The evolution of the China-U.S. trade negotiations remains a risk to our view as the outcome will weigh on investors’ expectations and sentiment. China’s Vs. DMs’ Credit Cycles Between 2009 and 2014, China’s credit cycle lagged the U.S. and EU’s broad money cycles (Chart 8). This counter-cyclicality is partly explained by its elevated level of exports to the U.S. and of hard goods to Europe. When the global economic cycle works in China’s favor – i.e., when the Fed and ECB are accommodative or fiscal stimulus is deployed in either or both regions – China’s exports rise as U.S. and EU aggregate demand increases. This typically reduces the need for endogenous fiscal or monetary stimulus within China. Chart 8China's Credit Cycle Lags U.S., EU Money Cycles On the other hand, when the global economic cycle contracts and fiscal and monetary policy ex China becomes a headwind, Chinese policymakers typically need to deploy fiscal and monetary policy to keep growth going, or at least avoid a contraction in their economy. Between 2016 and 2017, DM and China credit cycles aligned and increased simultaneously. Taking into account the 4-to-6-month lag between the time credit supply is increased and commodity demand rises, this created bullish conditions for metals and oil from 2H16 to 1H18, pushing copper prices up by 60%. In 2018, both regions’ cycles rolled over. Base metals markets currently are experiencing the consequences of this contraction in credit availability and tightening of financial conditions generally. Going forward, we expect China will step in to raise domestic demand and offset the impact of the decline in credit availability elsewhere, which is affecting demand for its exports in the short-term. In the medium-term, the U.S. and EU, along with India, do not appear to be inclined to absorb Chinese exports to the extent they did in the past, which means the pivot to domestically generated growth through consumer- and services-led demand is the most viable alternative Chinese policymakers have to keep growth on target. Bottom Line: The dovish turn of major DM central banks combined with a bottoming of China’s credit cycle will support cyclical commodities at the margin in the coming months. During the second half of this year, we expect a more significant pick up in China’s credit, setting the stage for a year-end rally in base metal prices. As a consequence, the impact of China’s credit growth on base metals demand could diminish compared to previous stimulus targeting industrial demand.   Hugo Bélanger, Senior Analyst Commodity & Energy Strategy HugoB@bcaresearch.com Robert P. Ryan, Chief Commodity & Energy Strategist rryan@bcaresearch.com   Footnotes 1      Please see “Russia’s oil companies ready to cut output until July: TASS,” published by reuters.com March 12, 2019. 2      Please see Fastmarkets MB’s Daily Steel, March 12, 2019. 3      Please see “Pervasive Uncertainty, Persuasive Central Banks,” published by BCA Research’s Global Fixed Income Strategy March 12, 2019. It is available at gfis.bcaresearch.com. 4      Please see “China Macro And Market Review,” published by BCA Research’s China Investment Strategy March 13, 2019. It is available at cis.bcaresearch.com. 5      See footnote 4 above. 6      For more details please see “EM: A Sustainable Rally Or A False Start?” published by BCA Research’s Emerging Market Strategy March 7, 2019. It is available at ems.bcaresearch.com. 7      Hamilton notes the HP filter can be problematic. In general, we agree with critics of the filter (i.e. it results in spurious dynamics that are unrelated with the true data-generating process, it has an end-point bias which affects its real-time properties, and it is highly dependent on the parameter selection). However, there are some arguments in support of using the HP filter to proxy the credit cycle. First, as long as there are no clear theoretical foundation for an accurate measurement of the credit cycle, empirical validation should remain the number one criteria by which one selects its proxy. Second, credit cycles vary in duration and this weakens the ability to construct a reliable proxy. The usual parameter used with the HP filter favors short-term cycles (i.e. ~ 2 years) while the Hamilton filter focuses on medium-term cycles (i.e. ~ 5 years). Therefore, both can convey useful information. Third, China’s aggregate credit variable in level has a quasi-linear trend and is roughly approximated by a trend-stationary process with breaks in the trend and constant. Such a process should converge in limit when decomposed using the HP filter. Please see James D. Hamilton (2018), “Why You Should Never Use the Hodrick-Prescott Filter,” The Review of Economics and Statistics, vol 100(5), pages 831-843. and Phillips, Peter C. B. and Jin, Sainan (2015), “Business Cycles, Trend Elimination, and the HP filter,” Cowles Foundation Discussion Paper No. 2005. 8      Granger causality refers to a statistical technique developed by Clive Granger, the 2003 Nobel Laureate in Economics, which is used to determine whether one variable can be said to have caused (or predicted) another variable, given the past performance of each. Using standard econometric techniques, Granger showed one variable can be shown to have “caused” another, and that two-way causality also can be demonstrated (i.e., a feedback loop between the variables can exist based on the historical performance of each). 9      Please see “Oil, Copper Demand Worries Are Overdone,” published by BCA Research’s Commodity & Energy Strategy February 14, 2019. It is available at ces.bcaresearch.com. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Trades Closed in 2019 Summary of Trades ​​​​​​​

Highlights Analysis on South Africa is published below. The “EM” label does not guarantee a secular bull market. None of the individual EM bourses has outperformed DM on a consistent basis over the past 40 years. EM share performance in both absolute terms and relative to DM has exhibited long-term cycles of around seven to 10 years. Getting these cycles right is instrumental to successful investing in EM. EM investing is predominantly about exchange rates. From a long-term (structural) perspective, EM equities are only modestly cheap in absolute terms but are very cheap versus the U.S. Feature We often receive questions from asset allocators about the long-term outlook for EM equities and currencies. The general perception among longer-term allocators is that while EMs may underperform over the short term, they always outperform developed markets (DM) in the long run. Consistently, the overwhelming majority of investors’ long-term return forecasts ascribe the highest potential return to EM equities and bonds among various regions and asset classes. This week we focus on the historical long-term performance of EMs. Contrary to popular sentiment, our findings show that EM stocks and currencies have not outperformed their U.S./DM peers in the past 40 years – as long as EMs have existed as an asset class. Hence, there is no guarantee that EM share prices and currencies will always outperform their DM counterparts on a secular basis going forward. Notably, EM share performance in both absolute terms and relative to DM has exhibited long-term cycles of around seven to 10 years. Getting these cycles right is instrumental to successful investing in EM. At the moment, the odds are that the current bout of EM equity and currency underperformance is not yet over, and more downside is likely before a major upturn emerges. The “EM” Label Does Not Guarantee A Secular Bull Market EM share prices have been in a wide trading range since 2010 (Chart I-1), despite the 10-year bull market in the S&P 500. Chart I-1Lost Decade For EM Stocks Remarkably, there is no single EM bourse that has been in a bull market during this decade (Chart I-2 and Chart I-3). This proves that this has indeed been a “lost” decade for EM. Chart I-2Individual EM Bourses: A Very Long-Term Perspective Chart I-3Individual EM Bourses: A Very Long-Term Perspective Historically, secular bull markets have been followed by bear markets not only in the boom-bust economies of Latin America, EMEA and Southeast Asia but also in former Asian tiger economies including Korea, Taiwan and Singapore (Chart I-4). This is despite the fact that per-capita real income has been growing rather rapidly in these Asian economies. Chart I-4Former Asian Tigers: Long-Term Equity Performance Remarkably, China and Vietnam have been exhibiting similar dynamics over the past 20 years – rapid per-capita real income growth and poor equity market returns (Chart I-5). Chart I-5China And Vietnam: Stock Prices And GDP Per Capita The message from all of these charts is as follows: Periods of industrialization and urbanization – even if successful – do not always entail structural bull markets. The U.S. fits this pattern as well. During the period between 1870 and 1900, the U.S. was experiencing industrialization and urbanization along with many productivity enhancements such as the steam engine, electricity and infrastructure construction. Even though America’s prosperity and real income per-capita levels surged during this period, corporate earnings per share and stock prices were rather flat (Chart I-6). Chart I-6The U.S. In The Late 1800s: Stocks, Profits And GDP Hence, rising per-capita real income and prosperity do not translate into higher share prices on a consistent basis. This is not to say that no country can ever deliver healthy stock market gains in the long run. Some certainly will, and it is our job to identify and expose these to clients. The point is that the “emerging market” status does not guarantee a structural bull market. Asset Allocation: Play Cycles Chart 7 illustrates that EM relative equity performance versus DM in general and the U.S. in particular has gone through several major swings over the past 40 years. Remarkably, none of the individual EM bourses has outperformed DM on a consistent basis over this time frame (Chart I-8A and I-8B). Chart I-7EM Versus DM: Relative Total Equity Returns Chart I-8ANo Single EM Bourse Has Outperformed DM In Past 40 Years Chart I-8BNo Single EM Bourse Has Outperformed DM In Past 40 Years Failure to outperform DM stocks is not only inherent for bourses in twin-deficit and inflation-prone regions/countries such as Latin America, Russia, Turkey, South Africa and South East Asia (including India), but it has also been true for share prices in rapidly growing countries such as China and Vietnam (Chart I-9). Chart I-9Chinese And Vietnamese Stocks Have Not Outperformed DM Remarkably, equity markets in the former Asian tigers – Korea, Taiwan and Singapore – have also failed to outperform their DM peers in the past 40 years (Chart I-10). This is in spite of the fact that real income per-capita growth in these Asian nations has by far outpaced that in both the U.S. and DM (Chart I-11). Chart I-10Former Asian Tigers Have Not Outperformed DM Equities... Chart I-11…Despite Economic Outperformance Evidently, the assumption that EM stocks will outperform DM equities on the back of higher potential growth rates is not validated by historical data. First, higher potential growth does not always ensure robust realized GDP growth. Second, even if real GDP-per-capita growth rises considerably, this does not always guarantee superior equity market returns. Some of the reasons for this include productivity benefits being transferred to employees rather than to shareholders, chronic equity dilution, and a misallocation of capital that boosts economic growth at the expense of shareholders. Bottom Line: EM relative stock performance versus DM has been fluctuating in well-defined long-term cycles. In our view, EM relative equity performance has not yet reached the bottom in this downtrend. We downgraded EM stocks in April 2010 and have been recommending a short EM equities / long S&P 500 strategy since December 2010 (please refer to Chart I-7 on page 5). EM Investing Is Primarily About Exchange Rates Exchange rates hold the key to getting EM equity cycles right for international investors. As demonstrated in Chart I-12, historically the bulk of EM equity return erosion has been due to currency depreciation. Chart I-12EM Investing Is All About Exchange Rates Exchange rates of structurally weak EM economies depreciate chronically. Common reasons include lack of productivity growth, high inflation, current account deficits, uncontrolled fiscal expansion, and reliance on volatile foreign portfolio flows. Periods of currency depreciation also occur in emerging Asian economies that have low inflation and typically run current account surpluses. Chart I-13 shows spot rates for Korea, Taiwan and Singapore versus the SDR which is a weighted average of USD, the euro, JPY, GBP, and CNY.1 Chart I-13Former Asian Tiger Currencies: Wide Fluctuations None of these Asian-tiger currencies has consistently appreciated versus the SDR. As in the case of share prices, there have been multi-year exchange rate swings. Further, U.S. dollar total returns on EM local bonds are also primarily driven by their currencies (Chart I-14). Consequently, the cycles in EM local currency bonds match EM exchange rate cycles. Chart I-14Total Return On Local Currency Bonds EM credit spread fluctuations are also by and large contingent on their exchange rates. Credit spreads on EM sovereign and corporate U.S. dollar bonds gauge debt servicing risk. The latter is highly influenced by exchange rates. Currency depreciation (appreciation) increases (decreases) debt servicing costs thereby affecting credit spreads. Bottom Line: Exchange rate fluctuations are driven by macro crosscurrents, making macro an indispensable know-how for EM investing. We maintain that EM currencies are susceptible to renewed weakness against the U.S. dollar as China’s growth continues to weaken, weighing on EM growth and thereby their respective exchange rates (Chart I-15). In turn, the U.S. dollar is a countercyclical currency and does well when global growth decelerates. Chart I-15EM Currencies Are Pro-Cyclical Valuations: The Starting Point Matters… In recent years, a long-term bullish case for EM equities and currencies has often been made on the grounds of cheap valuations. Chart I-16 illustrates the equity market-cap weighted real effective exchange rate for EM ex-China, Korea and Taiwan – a measure that is pertinent for both EM equity and fixed-income investors.2 It reveals that EM currency valuations are only slightly below their historical mean. Chart I-16EM Ex-China, Korea, Taiwan Currencies Are Modestly Cheap As to the CNY, KRW and TWD, their valuations are not at an extreme, and the CNY holds the key. The main long-term risk to the RMB is capital outflows from Chinese households and companies as discussed in February 14 report. For long-term investors, the pertinent equity valuation yardstick is the cyclically adjusted P/E (CAPE) ratio. The idea behind the CAPE model is to remove cyclicality of corporate profits when computing the P/E ratio – i.e., to look beyond a business cycle. Hence, the CAPE ratio is a structural valuation model – i.e., it works in the long term. Only investors with a time horizon greater than three years should use this valuation measure in their investment decisions. Our CAPE model gauges equity valuations under the assumption of per-share earnings converging to their trend line. The latter is derived by a regression of the cyclically adjusted EPS in real U.S. dollar terms on time. The EM CAPE ratio presently stands at 0.5 standard deviations below its historical mean (Chart I-17). This means EM stocks are modestly cheap from a long-term perspective. Meanwhile, the U.S.’s CAPE ratio is very elevated (Chart I-18). Chart I-17EM Equities Are Modestly Cheap From AA1 Structural Perspective Chart I-18U.S. Stocks Are Expensive From AA1 Structural Perspective On a relative basis, EMs are very attractive relative to U.S. stocks (Chart I-19). This entails that the probability of EM stocks outperforming U.S. equities is very high from a secular perspective – longer than three years. Chart I-19EM Equities Are Cheap Versus U.S. From AA1 Structural Perspective Nevertheless, a caveat is in order. Our CAPE model assumes that EPS in real U.S. dollar terms will rise at the same pace as it has historically. The slope of the time trend – the historical compound annual growth rate (CARG) of EPS in inflation-adjusted U.S. dollar terms – is 2.8% for EM and 2% for the U.S. Please note that we determined the earnings time trend (trend line) using historical ranges – 1983 to present for EM, and 1935 to present for the U.S. Hence, these CAPE models assume that EM EPS will grow 0.8 percentage points (2.8% minus 2%) faster than U.S. corporate EPS in inflation-adjusted U.S. dollar terms, as they have done historically. Under this assumption, EM stocks are considerably cheaper than the U.S. market. That said, in the medium term, corporate earnings are the key driver of EM share prices, and contracting profits pose a risk to EM performance, as discussed in our February 21 report. Bottom Line: From a long-term perspective, EM equities and currencies are only modestly cheap in absolute terms. Based on our CAPE ratio model, EM stocks are very cheap versus the U.S. However, the CAPE ratio is a structural valuation measure, and only investors with a time horizon of longer than three years should put considerable emphasis on it. …But Beware Of A Potential Value Trap If for whatever reason there is a change in the slope of the EM EPS long-term trend – i.e., per-share earnings fail to expand in the coming years at their historical rate, as discussed above, our CAPE model would be invalidated. In such a case, EM share prices are unlikely to enter a secular bull market in absolute terms and outperform their U.S. counterparts structurally. The key to sustaining the current upward slope in the long-term trajectory of EPS in real U.S. dollar terms is for EM/Chinese companies to undertake corporate restructuring and increase efficiency. Critically, recurring Chinese credit and fiscal stimulus as well as cheap and abundant money from international investors have not fostered corporate restructuring in China, nor in other EM countries. The basis is that easy and cheap financing and economic growth propped-up by periodic Chinese stimulus has made companies complacent, undermining their productivity and efficiency. The ultimate outcome will be weak corporate profitability over the long run. Another long-term risk to corporate earnings in China and some other EMs is the expanding role of the state in the economy. In these circumstances, China/EM corporate profitability will also suffer over the long run. The basis is that in any country the private sector is better than the government in generating strong corporate earnings. Bottom Line: Without structural reforms and corporate restructuring in EM/China, EM stocks are unlikely to outperform their DM peers on a secular basis. Investment Conclusions The medium-term EM outlook remains poor for the reasons we elaborated on in last week’s report titled, EM: A Sustainable Rally or A False Start? Further, investor sentiment on EM is very bullish, and positioning in EM equities and currencies is elevated (Chart I-20). We continue to recommend underweighting EM stocks, credit markets and currencies versus their DM counterparts and the U.S. in particular. Chart I-20Investors Are Very Bullish On EM From a long-term perspective, EM equity and currency valuations are modestly cheap. However, a durable long-term expansion in EM economies is contingent on a sustainable bottom in Chinese growth. The latter hinges on deleveraging and corporate restructuring in China, neither of which have occurred to a meaningful extent. For EM equity portfolios, we presently recommend overweighting Mexico, Brazil, Chile, central Europe, Russia, Thailand and Korean non-tech stocks. Our current (not structural) underweights are South Africa, Indonesia, India, the Philippines, Hong Kong and Peru. Within the EM equity space, two weeks ago we booked triple-digit profits on our strategic long positions in EM tech versus both the overall EM index and EM materials stocks, respectively. These positions were initiated in 2010. The basis for these strategic recommendations was our broader theme for the decade of being long what Chinese consumers buy, and short plays on Chinese construction, which we initiated on June 8, 2010. This week we are closing our long central European banks / short euro area banks equity position. We recommended it on April 6, 2016, and it has produced a 14% gain since then. Arthur Budaghyan Chief Emerging Markets Strategist arthurb@bcaresearch.com South Africa: Debt Deflation Or Currency Depreciation? South Africa’s public debt dynamics are on an unsustainable track. Two prerequisites for public debt sustainability are (1) for interest rates to be below nominal GDP growth or (2) continuous robust primary fiscal surpluses. Hence, a government can stabilize its debt-to-GDP ratio by either having nominal GDP above its borrowing costs, or by running persistent and sizable primary fiscal surpluses. Neither of these two stipulations are presently satisfied in South Africa. The gap between government local currency bond yields and nominal GDP growth is at its widest in over the past 10 years (Chart II-1). Meanwhile, the primary fiscal deficit is 0.75% of GDP (Chart II-2). Chart II-1South Africa: An Unsustainable Gap Chart II-2South Africa Has Not Had A Primary Fiscal Surplus In A Decade Faced with very low real potential GDP growth stemming from the economy’s poor structural backdrop, the authorities in South Africa ultimately have two choices to stabilize the public debt-to-GDP ratio: Tighten fiscal policy substantially, trying to achieve persistent large primary budget surpluses; or Inflate their way out of debt, which would require a large currency depreciation to boost nominal GDP growth above borrowing costs. With this in mind, we performed a simulation on public debt, assuming fiscal tightening but no substantial currency depreciation (Table II-1). The first scenario uses the 2019 consolidated budget government assumptions and projections for nominal GDP, government revenues and expenditures, i.e., it is the government's scenario. In this scenario, the public debt-to-GDP ratio rises only to 58% by the end of the 2021-‘22 fiscal year. However, government forecasts always end up being optimistic. We believe this scenario is implausible due to its overestimation of nominal GDP, and hence government revenue growth. As the government tightens fiscal policy, nominal GDP growth and ultimately government revenue will disappoint substantially. For the second scenario, we used government projections for fiscal spending in the coming years, but our own estimates for nominal GDP and government revenue growth. Notably, excluding interest payments and fiscal support for ailing state-owned enterprises like Eskom, nominal growth of government expenditures in the current year is at 7.5%, and estimated to be 6.8% the next two fiscal years. That is why we project nominal GDP and government revenue growth to be very weak. The basis of our assumption is as follows: Barring considerable currency depreciation, as the authorities undertake substantial fiscal tightening in the next three years, nominal GDP and consequently government revenue growth will plunge. Importantly, government revenues exhibit a non-linear relationship with nominal GDP – government revenues fluctuate much more than nominal GDP (Chart II-3). Chart II-3Government Revenues Are 'High-Beta' On Nominal GDP Growth As government revenue growth underwhelms, the primary deficit will widen and the public debt-to-GDP ratio will escalate, reaching 70% of GDP by the end of the 2021-‘22 fiscal year, according to our projections (Table II-1). Overall, without considerably lower interest rates and material currency depreciation, the government’s financial position will enter a debt deflation spiral. Fiscal tightening will hurt nominal growth damaging fiscal revenues. As a result, the fiscal deficit will widen – not narrow – and the debt-to-GDP ratio will rise. Therefore, the only feasible option for South Africa to stabilize public debt is to reduce interest rates dramatically and depreciate the currency. This will engender higher inflation and nominal growth, thereby boosting government revenues and capping the public debt burden. At 10%, the share of foreign currency debt as part of South Africa’s public debt is low. Hence, currency depreciation will do less damage to public debt dynamics than keeping interest rates at high levels. On the whole, the rand is a very structurally weak currency, and is bound to depreciate due to deteriorating public debt dynamics. Chart II-4 plots the real effective exchange rate of the rand based on CPI and PPI. It is evident that its valuation is not yet depressed. Chart II-4The Rand Is Modestly Cheap Meanwhile, cyclical headwinds also warrant currency depreciation (Chart II-5). Chart II-5Widening Trade Deficit Warrants Currency Depreciation Market Recommendations Continue shorting the ZAR versus the U.S. dollar and the MXN. Consistent with the negative outlook for the exchange rate, investors should underweight South African local currency government bonds and sovereign credit within respective EM portfolios. Finally, we recommend EM equity portfolios remain underweight South African equities. Andrija Vesic, Research Analyst andrijav@bcaresearch.com Arthur Budaghyan Chief Emerging Markets Strategist arthurb@bcaresearch.com   Footnotes 1       Special Drawing Rights. The value of the SDR is based on a basket of five currencies: the U.S. dollar, the euro, the Chinese renminbi, the Japanese yen, and the British pound sterling. 2      We exclude these three currencies since their bourses have very large equity market cap in the EM stock index and, hence, would make any aggregate currency measure unrepresentative for the rest of EM.   Equity Recommendations Fixed-Income, Credit And Currency Recommendations