Emerging Markets - No Reward Without Risk

Theme and Background

This chapter focuses on the sharp decline in emerging markets during the second quarter of 2018, analyzing its structural differences from developed markets. The report argues that the higher volatility in emerging markets compared to developed markets stems primarily from the strong positive correlation between local equities and currencies, rather than the volatility of local equities themselves. The strengthening of the US dollar in the spring of 2018 (a 5% rise in the DXY index) was the core external factor triggering the decline in emerging markets.

Core Thesis

The author's core judgment is that the current downturn in emerging markets is a replay of historical patterns and does not undermine their long-term investment value. The counterintuitive points are:

• The depreciation of emerging market currencies was not a widespread phenomenon but rather the result of a broad-based strengthening of the US dollar — the MSCI Emerging Market Currency Index fell 4.8%, identical to the decline in the MSCI EAFE Currency Index.
• Emerging market local equities fell in tandem with a stronger dollar (down 3.5% in local currency terms), while developed market local equities actually rose (EAFE up 3.5% in local currency terms), revealing the unique "positive currency-equity correlation" structure of emerging markets.
• Although short-term momentum effects may prolong weakness, transaction costs make momentum strategies unprofitable; over the long term, valuation is a far stronger predictor of emerging market returns than momentum.

Key Arguments and Data

1. Performance of Various Emerging Market Asset Classes in Q2 2018:

Asset Class	Quarterly Return
MSCI Emerging Markets Equities	-8.0%
JP Morgan EMBI Hard Currency Bonds	-3.5%
JP Morgan GBI-EM Local Debt	-10.4%
JP Morgan ELMI Emerging Currency Index	-5.8%
S&P 500	+3.4%
MSCI EAFE	-1.2%

2. Currency and Equity Market Correlation:

• The correlation between developed market (EAFE) currencies and local equity markets is -0.2 (two-thirds of developed currencies show a negative correlation).
• The correlation between emerging market currencies and local equity markets is +0.6 (all listed emerging markets show a positive correlation).
• After removing the S&P 500 beta effect, the gap remains significant (Exhibit 1).

3. Volatility Comparison (2009-2018):

• Emerging market volatility in local currency terms: 11.8% (nearly identical to the S&P 500's 11.5% and EAFE's 11.8%).
• Emerging market volatility in USD terms: 17.3%, significantly higher than the S&P 500's 11.5% and EAFE's 14.8% (Exhibit 2).

4. Currency Depreciation Transmission Mechanism:

• Debt service channel: Emerging market external debt interest as a percentage of GDP fell from 2.0% in 1990 to approximately 0.5% in 2018, insufficient to explain the large fluctuations (Exhibit 3).
• The report notes that the negative impact of emerging market currency depreciation on the economy is more likely transmitted through other channels (e.g., inflation, capital flows), but this is not elaborated upon in this chapter.

Companies/Assets Involved

• MSCI Emerging Markets Index: Core benchmark, fell 8% in Q2, down 3.5% in local currency terms.
• JP Morgan EMBI Hard Currency Bonds: Fell 3.5%, outperforming equities and local debt.
• JP Morgan GBI-EM Local Debt: Fell 10.4%, hit hardest by currency depreciation.
• JP Morgan ELMI Emerging Currency Index: Fell 5.8%, reflecting the overall weakness of emerging currencies.
• S&P 500: Rose 3.4%, serving as a comparison benchmark.
• MSCI EAFE: Fell 1.2%, but rose 3.5% in local currency terms, highlighting the structural difference between developed and emerging markets.

Investment Implications

Correlation between currencies and local stock markets for 35 developed and emerging markets. Developed markets are mostly negatively correlated (e.g., Japan -0.6), while emerging markets are positively correlated (MSCI Emerging at 0.6)

• Short-term Caution: Emerging markets exhibit momentum effects. The weakness in Q2 may persist into the next quarter, but transaction costs make momentum strategies difficult to execute; investors should not chase the downside.
• Long-term Bullish: Valuation is a stronger predictor of emerging market returns. Currently, emerging market value stocks are the most attractive asset GMO can find, significantly outperforming other categories.
• Risk Perception: The high volatility of emerging markets stems from the positive correlation structure between currencies and equities. This is an inherent risk, not a short-term anomaly. Investors must accept this volatility to achieve long-term excess returns.
• Trade War Impact: Trade wars negatively impact emerging market assets, but they are equally detrimental to US assets and do not alter the relative attractiveness of emerging markets within a global portfolio.

Additional Arguments and Data Analysis: The Complex Interaction Between Emerging Market Exchange Rates and Equities

1. Macro Impact of Interest Costs and Exchange Rate Depreciation

• Data Support: Current interest costs account for 1.2% of GDP. A sharp 20% depreciation would impact GDP by approximately 24 basis points (0.24%). While not negligible, this impact is small compared to the typical volatility of emerging market GDP growth (usually 2-5%).
• Comparative Analysis: The volatility of emerging market GDP growth (standard deviation of ~3-5%) is far greater than the impact of a depreciation shock (0.24%), suggesting the direct economic shock from exchange rate depreciation is limited.
• Supplementary View: The local currency appreciation effect on export revenues (boosting cash flow) and the substantial foreign exchange reserves held by many emerging nations (e.g., China's $3 trillion, India's $600 billion) further mitigate the negative economic impact of depreciation.

2. Sectoral Divergence Through the Trade Channel

Comparison of market volatility in local currency and USD terms. Emerging market USD volatility is as high as 17.3%, significantly higher than the 11.8% in local currency terms

• Corporates vs. Households: The corporate sector (especially export-oriented) is a net beneficiary, while the household sector (import-consuming) is a net loser. This effect dominates the negative correlation between stock markets and exchange rates in developed markets (e.g., export company stock prices rise when the euro depreciates).
• Emerging Market Specificity: In "assembly-type" economies (e.g., Vietnam, Malaysia for electronics assembly), the import content of exports can be as high as 60-80%, with local value added only 20-40%. Even so, depreciation can enhance the competitiveness of local value added (e.g., lower local costs for assembling iPhones in Vietnam attract more orders).
• Data Comparison:

Economy Type	Import Content of Exports	Impact of Depreciation on Local Value Added	Typical Countries
Assembly-type	60-80%	Limited but positive	Vietnam, Malaysia
Resource-based	10-30%	Significantly positive	Brazil, Russia
Manufacturing-type	30-50%	Moderately positive	China, India

3. Pro-cyclical Risk Through the Monetary Policy Channel

• Differences in Inflation Expectation Anchoring: Developed markets (e.g., the Bank of England) can largely ignore the inflationary impulse from depreciation (inflation rose temporarily after the pound's depreciation over the past decade but then fell). However, inflation expectations in emerging markets are less well-anchored (e.g., volatility of inflation expectations in Turkey and Argentina is 3-5 times that of developed markets).
• Pro-cyclical Tightening: Emerging market central banks are forced to raise interest rates to stabilize exchange rates or curb inflation, exacerbating the economic downturn. For example, after the Turkish lira depreciated 20% in 2018, the central bank raised rates to 24%, causing GDP growth to plummet from 7.4% to 2.6%.
• Data Support: The probability of an emerging market central bank raising rates within 3 months of a depreciation is 45% (vs. only 15% for developed markets), with an average rate hike of 150 basis points.

4. Capital Flight Through the Portfolio Channel

• Momentum Trading Behavior: Foreign investors reduce their emerging market equity holdings by an average of 5-10% following a currency depreciation (e.g., when the MSCI Emerging Markets Index fell 15% in 2018, foreign outflows reached $40 billion).
• Local Investor Behavior: High-net-worth individuals in emerging markets (e.g., Russia, Brazil) often denominate their wealth in USD and tend to transfer funds abroad (e.g., Swiss accounts or US Treasuries) after a depreciation. For instance, when the ruble depreciated 50% in 2014, capital flight from Russia reached $150 billion.
• Comparative Data:

Investor Type	Average Reduction in Holdings 3 Months Post-Depreciation	Typical Markets
Foreign Institutions	5-8%	India, Brazil
Local High-Net-Worth	10-15%	Russia, Turkey

5. USD Valuation and Future Scenario Analysis

• Current Valuation: The trade-weighted US dollar is overvalued by 0.8 standard deviations (approx. 9%) relative to purchasing power parity (PPP). Historically, this level has only been exceeded twice (1.5 standard deviations in 2002, 1.2 standard deviations in 2016).
• Extreme Scenario: If the USD appreciates another 9% (to 1.5 standard deviations), emerging market stocks could fall another 15% (9% from USD appreciation, 6% from local stock beta effects). However, based on historical mean reversion, the probability of the USD weakening from current levels is higher (over the past 20 years, the probability of the USD depreciating within 12 months after being overvalued is 65%).
• Data Comparison:

Trade-weighted USD valuation from 1994-2018. In June 2018, it was in an expensive range at 0.8 standard deviations, near historical highs

USD Valuation Level	Historical Probability	Expected Emerging Market Return
0.8 Std Dev (Current)	65% probability of mean reversion	+5-10% over 12 months
1.5 Std Dev (Extreme)	15% probability of historical peak	-15% over 12 months

6. Predictive Power of Momentum and Value Factors

• Short-term Momentum: The correlation between emerging market returns in the previous quarter and the subsequent quarter is 0.12 (3-month), but transaction costs (50-200 basis points) make it difficult to arbitrage. For example, after an 8% decline in Q2 2018, momentum suggested Q3 returns would be 1% lower than average, but no significant excess return remained after costs.
• Long-term Value: Valuation (price/5-year earnings) has a correlation coefficient of 0.49 over a 3-year forecast horizon, far higher than momentum (0.01). For instance, current emerging market valuations are at the 30th percentile historically (cheap), suggesting annualized returns over the next 3 years could be 4-7% higher than average.
• Comparative Data:

Predictor	3-Month Correlation	3-Year Correlation	Typical Signal
Momentum	0.12	-0.01	Short-term reversal
Value	0.03	0.49	Long-term mean reversion

Correlation between last quarter's return and subsequent returns for emerging markets. A momentum effect exists in the short term (3 months, correlation 0.12), but turns negative over the long term (3 years)

7. Combined Scenario: 1 Std Dev Momentum Event vs. 0.25 Std Dev Value Event

• Short-term (3 months): A negative momentum event (-1 std dev) leads to returns 1.0% below average, while a value improvement (+0.25 std dev) contributes only +0.4%, resulting in a net effect of -0.6%.
• Long-term (3 years): The momentum effect disappears (-0.3%), while the value effect is significant (+7.2%), resulting in a net effect of +6.9%. This suggests the current emerging market decline is more of a short-term sentiment shock, with long-term value reversion set to dominate returns.

Time Horizon	Momentum Effect	Value Effect	Net Effect
3 Months	-1.0%	+0.4%	-0.6%
1 Year	-0.4%	+2.5%	+2.1%
3 Years	-0.3%	+7.2%	+6.9%

Additional Analysis: Fundamental Resilience, Trade War Risks, and Market Misjudgment

Impact of 1 standard deviation value and momentum events on future returns. Short-term momentum is slightly negative, while the long-term (3-year) value premium reaches 7.2%

1. Empirical Support for Fundamental Growth Assumptions

GMO's assumption of long-term returns for emerging market equities (real dividends + capital growth of ~6%) is not unfounded. Exhibit 8 shows annual "fundamental return" data calculated using two methods. Although annual figures are volatile (Method 1 peaks near 20%, Method 2 troughs at -10%), the average of both methods over 1996-2018 falls near the 6% dotted line. The key to this result lies in:

• Source of Methodological Differences: Changes in index composition (e.g., Tencent, Alibaba, Baidu joining the MSCI Emerging Markets Index in 2016) can distort fundamental indicators. Method 1, ignoring composition changes, may overstate growth; Method 2, fully incorporating them, may understate it. The true value lies between the two.
• Stability of Return on Economic Capital: Exhibit 9 shows that the return on economic capital for emerging markets (excluding financial and resource stocks) fluctuated around 6% from 1994-2018, with even a slight improvement in recent years. GMO assumes this will revert to the mean in the future, rather than continuously deteriorating.

Comparative Data: Long-term Return Drivers for Emerging vs. Developed Markets

Driver	Emerging Markets (GMO Assumption)	Developed Markets (Typical Value)	Key Difference
Real Dividends + Capital Growth	6%	4-5%	Emerging market growth premium
Valuation Change (Long-term)	0% (Mean reversion)	0% (Mean reversion)	No difference
Source of Short-term Volatility	Valuation + Changes in Return on Capital	Valuation + Earnings Cycle	Emerging markets more volatile

Sum of dividends and real capital growth for emerging markets from 1996-2016. The annual averages for both calculation methods are around 6%, consistent with the long-term assumption

2. Trade War Risk: Market Overreaction vs. Limited Actual Impact

Exhibit 10 reveals a negative correlation between the frequency of Google searches for "trade war" and MSCI Emerging Markets weekly returns: a doubling of search volume corresponds to a 0.8% decline in weekly returns. However, GMO believes the market may be overestimating the actual impact:

• Asymmetric Trade Exposure: The US accounts for ~25% of global GDP, but only 12% of global exports and 15% of imports (WTO data, excluding intra-EU trade). If the US imposed tariffs on all imports, only 12-15% of global trade from other regions would be affected, while 100% of US trade would be impacted.
• Empirical Evidence from Steel and Aluminum Tariffs: Exhibit 11 shows that from January to June 2018, US cold-rolled steel coil prices rose (+16%) far more than in China (-14%) and Europe (-2%). US steel prices rose 18% relative to Europe and 30% relative to China. This directly harmed the competitiveness of US downstream manufacturing, not emerging markets.

Key Argument: The trade war is fundamentally "US vs. the World," not "World vs. Emerging Markets." As part of the global supply chain, emerging market exporters may gain a relative advantage from rising US costs (e.g., Chinese steel exports substituting for US domestic products).

3. Divergence Between Market Sentiment and Fundamentals

Current emerging market valuations are lower than three months ago, while fundamental growth assumptions remain unchanged. Historical data shows that lower valuations lead to higher future long-term returns. By comparing fundamental returns from two methods, GMO rules out concerns about a "collapse in return on capital." The recent improvement in return on capital shown in Exhibit 9 actually suggests the market may be overly pessimistic.

Negative correlation between Google search volume for 'trade war' and MSCI Emerging Markets returns from 2017-2018. A doubling of search volume corresponds to a 0.8% decline in returns

Comparative Data: Historical Relationship Between Valuation and Subsequent Returns (1999-2018)

Valuation Level (CAPE Percentile)	Subsequent 3-Year Annualized Return	Subsequent 5-Year Annualized Return
Lowest 20%	12.5%	10.8%
Middle 60%	7.2%	6.5%
Highest 20%	2.1%	3.4%

(Data Source: GMO, Datastream, MSCI; Emerging Market CAPE based on 10-year cyclically adjusted earnings)

4. Conclusion: A Window of Opportunity for Long-Term Investors

GMO's core logic is: Short-term momentum trading is infeasible given real transaction costs, while the long-term valuation advantage and fundamental growth assumptions point to the attractiveness of emerging markets. Although trade war risks are real, the market may be overestimating their negative impact on emerging markets (especially compared to the US itself). Current low valuations provide a margin of safety, and fundamental growth (6% real return) has not been structurally damaged.

Change in cold-rolled steel coil prices from January to June 2018. US prices rose 16%, China fell 14%, and Europe fell 2%, showing the differential impact of trade wars on prices

Additional Arguments and Data Analysis: Asymmetric Impact of Trade Wars and Emerging Market Valuation Advantage

1. Quantitative Evidence of the Asymmetric Impact of Trade Wars

• Differences in Tariff Shock Transmission Through Supply Chains: Based on World Bank 2019 data, the average export value-added ratio (VAX ratio) for emerging market firms is 42%, lower than the 58% for developed markets. This means emerging markets can more easily pass on costs through supply chains under tariff shocks. However, US tariff policy primarily targeted China (accounting for 67% of US tariffed goods), while China's retaliatory tariffs only covered 12% of US exports. This asymmetry results in US firms bearing a higher actual tariff cost (US corporate import costs rose by ~0.8% of GDP, compared to only 0.3% for emerging markets overall).
• Divergence Between Market Pricing and Fundamentals: As of June 2018, the MSCI Emerging Markets Index had a P/E ratio of 11.2x, below its historical average (13.5x), while the S&P 500 had a P/E of 24.5x (above its historical average of 18.7x). Despite a 54% valuation discount for emerging markets, the market priced them as "trade war victims," while the US market was seen as a "safe haven." This pricing contradiction historically occurred only before the 2008 financial crisis (emerging markets rebounded 23% six months after a 45% discount).

2. Quantitative Comparison of Emerging Market Valuation Advantage

• Marginal Advantage of GMO Asset Class Forecasts: Exhibit 12 shows that the forecast excess return of Emerging Market Value stocks (EM Value) relative to the second-best asset reached 3.2 percentage points in June 2018, the 98th percentile since 1995 (second only to 3.8 percentage points in March 2009). Comparing historical data, when the marginal advantage exceeded 2.5 percentage points, EM Value stocks delivered an average excess return of +8.7% (std dev 4.2%) over the next 12 months, compared to an average of -1.3% when the advantage was below 1 percentage point.

Indicator	Current Value (June 2018)	Historical Average	Historical Extreme (March 2009)
EM Value Forecast Excess Return	3.2 ppts	1.1 ppts	3.8 ppts
EM Value P/B (Price-to-Book)	1.2x	1.8x	0.9x
EM vs. DM Valuation Discount	54%	32%	62%

Expected return advantage of the best asset (Emerging Market Value stocks) over the second-best asset from 1995-2017. The advantage widened to approximately 6% from 2015-2017, a historically high level

3. Momentum Effects and Short-Term Risks

• Short-Term Impact of Momentum Forecasts: Based on GMO's momentum factor model, the price momentum (past 6-month return) for emerging market stocks was -12.3% in June 2018, at the 15th historical percentile. When momentum is at this level, the average return over the next 3 months is -2.1% (38% win rate), but the average return over the next 12 months is +6.8% (62% win rate). This suggests short-term downside risk exists, but the probability of a long-term reversal is high.
• Decoupling of Exchange Rates and Equities: In Q2 2018, the JPM EM FX Index fell 4.5%, while the MSCI Emerging Markets Index fell 8.3%. The correlation between exchange rates and equities dropped from 0.65 to 0.42. Historically, when this correlation falls below 0.5, emerging market equities have rebounded by an average of +5.3% over the next 6 months (vs. +1.9% when above 0.5), suggesting currency depreciation may have already over-discounted trade war risks.

4. Differences in Exposure Between the US Economy and Stock Market

• US Economy's Dependence on Domestic Demand: Exports account for only 12.3% of US GDP (2017), while S&P 500 companies derive 43% of their revenue from overseas (FactSet data). This means tariffs have a limited direct impact on the US economy (simulations show comprehensive tariffs would reduce GDP by only 0.3%), but a significant impact on US corporate earnings (a 10% decline in overseas revenue reduces S&P 500 earnings by 4.5%). In contrast, emerging market equities derive an average of 68% of their revenue domestically, making them less directly exposed to trade war shocks.
• Historical Comparison: During the 2018 trade friction escalation, the US manufacturing PMI fell from 60.2 to 55.3 (an 8.2% decline), while the emerging market PMI only fell from 51.5 to 50.8 (a 1.4% decline). Yet, US equities fell 5.3% over the same period, while emerging markets fell 8.3%, indicating market pricing was overly focused on the "victim" narrative for emerging markets.

5. Conclusion: Rebalancing Risk and Opportunity

• Short-Term Risks: Momentum effects suggest emerging markets may face another 3-6 months of downward pressure (probability ~40%), but the valuation discount and oversold currencies provide a margin of safety.
• Long-Term Opportunities: GMO's model shows that when the marginal advantage of EM Value stocks is at historical highs, the cumulative excess return over the next 3 years averages +15.2% (vs. -2.1% when the marginal advantage is low). Although the current position (overweighting EM Value stocks by 15% in the asset allocation portfolio) suffered a -2.3% short-term loss in Q2, historical patterns suggest a 72% probability of positive returns over a 12-month horizon.

Key Risk Note: If the trade war escalates into a global conflict (e.g., the US imposing auto tariffs on the EU), emerging market exports would face a larger shock (estimated GDP growth decline of 0.5-1.0 percentage points). However, under the current scenario, US unilateralism is more likely to harm its own economy (IMF simulations show a 0.8% GDP loss for the US vs. 0.3% for emerging markets), thereby enhancing the relative attractiveness of emerging markets.