South Korea

The South Korean economy is facing strong deflationary pressures, requiring significant and additional rate cuts. Meanwhile, 10-year government bonds yield are still at 1.4%, 75 basis points over 10-year US Treasurys (Chart II-1). Hence, Korea’s bond yields offer good value for fixed-income investors and have considerable downside. We have been receiving 10-year swap rates in Korea since 2011 and are reiterating this recommendation: Chart II-2 shows that the GDP deflator has been negative since 2018, and core and trimmed mean consumer prices are flirting with deflation. Chart II-1Korean Government Bonds Yields: More Room To Fall Korean Government Bonds Yields: More Room To Fall Korean Government Bonds Yields: More Room To Fall Chart II-2The Korean Economy Is Flirting With Deflation The Korean Economy Is Flirting With Deflation The Korean Economy Is Flirting With Deflation   Falling prices amid elevated corporate and household debt levels – at 102% and 96% of GDP respectively – is toxic. The basis is price deflation increases real debt burdens. Notably, the debt service ratio for businesses and households is very high at 19.9% of GDP. Exports – which account for some 40% of GDP – are plunging. The business survey from Bank of Korea suggests exporters’ business sentiment plunged by a record in May and is close to 2008 levels, pointing to a dreadful export outlook. (Chart II-3) Domestic demand will remain weak, despite the large fiscal response to the COVID-19 outbreak. Business investment and hiring will be depressed for a while, undercutting consumer spending (Chart II-4). Chart II-3Exports In Freefall Exports In Freefall Exports In Freefall Chart II-4Less Investment Plan And Poor Employment Outlook Less Investment Plan And Poor Employment Outlook Less Investment Plan And Poor Employment Outlook   Finally, residential investment was in the doldrums even before the COVID-19 outbreak. Chart II-5 illustrates that declining residential construction permits preclude lower residential construction for the rest of the year. The Bank of Korea will have to cut interest rates considerably this year. From a big-picture perspective, there is no reason why Korea’s policy rate should not be reduced close to zero as is the case in advanced economies. Korea’s economy shares many similarities with advanced economies like high debt levels and persistent deflationary pressures. On top of this, Korea is much more exposed to global trade, which makes its cyclical outlook worse, heralding substantial monetary easing. Exchange Rate Low interest rates could undermine the Korean won, even though the exchange rate has not historically been driven by interest rate differentials. The key driver of the won – shrinking global trade volumes and deflating tradable goods prices – warrants a cheaper currency to mitigate the negative impact on corporate profitability (Chart II-6). Chart II-5Falling Residential Construction Permits Falling Residential Construction Permits Falling Residential Construction Permits Chart II-6Deflating Export Prices Herald Currency Depreciation Deflating Export Prices Herald Currency Depreciation Deflating Export Prices Herald Currency Depreciation   Besides, deflation in DRAM prices (Chart II-7) as well as DRAM sales point to further currency depreciation and lower Korean tech stock prices (Chart II-8). Chart II-7Deflating Semiconductor Prices... Deflating Semiconductor Prices... Deflating Semiconductor Prices... Chart II-8Semiconductor Prices Are Still Deflating ...Does Not Bode Well For Tech Stocks Semiconductor Prices Are Still Deflating ...Does Not Bode Well For Tech Stocks Semiconductor Prices Are Still Deflating ...Does Not Bode Well For Tech Stocks     Overall, a weak currency is needed to alleviate deflationary pressures currently present in the economy. Stocks We are negative on the KOSPI in absolute terms but continue to recommend that EM-dedicated equity portfolio investors overweight this bourse. Despite being a highly cyclical market, we believe the KOSPI’s outperformance will be due to its large weight in tech stocks. The latter will benefit from China’s ambitious tech-related infrastructure plan in the coming years. The plan includes construction of Information Transmission, Software and Information Technology Services, such as 5G networks, industrial internet and data centers. We expect total investment will reach between US$182 billion and $266 billion by the end of 2020, an increase of 30-50% over last year. Importantly, 40% of Korea’s semiconductor exports are purchased by China. We have been playing the semiconductor theme via Korea rather than Taiwan because the latter is a wild card amid escalating geopolitical tensions between the US and China. Our geopolitical team expects a flare up in US-China tensions ahead of US elections this year, and Taiwan could become one of the focal points. Bottom Line: Continue receiving 10-year swap rates, shorting the won against the US dollar and overweighting the KOSPI within an EM dedicated equity portfolio. Lin Xiang, CFA Research Analyst linx@bcaresearch.com

Feature Analysis on Korea & South Africa are available on pages 6 and 10, respectively. Mexico: Balancing Pros And Cons We have been overweight Mexican sovereign credit and local currency bonds as well as equities relative to the respective EM benchmarks. Our rationale for this stance has been the fact that Mexico’s macro risk premium relative to other EMs has been, in our opinion, wider than it should have been. However, the COVID-19 outbreak has introduced new dimensions into this analysis. On one hand, there are a number of positives that still warrant a lower macro risk premium on Mexican assets: The nation’s public debt burden is rising sharply but is not yet at an unsustainable level. We estimate that assuming (1) a nominal GDP contraction of 7% in 2020, (2) an overall fiscal deficit of 4.7% of GDP this year, and (3) the peso’s exchange rate versus the US dollar at 26, the gross public debt-to-GDP ratio will rise to 49% from 37% currently (Table I-1). If we assume the government takes over all SOE debt, including that of Pemex, total gross public debt will rise to 62% of GDP (Table I-1). While non-trivial, Mexico’s public debt burden is considerably lower than those in large EM countries like Brazil and South Africa. Table I-1Mexico's Public Debt Burden Mexico, Korea & South Africa Mexico, Korea & South Africa Chart I-1Mexico: Real And Nominal Rates Are Too High Mexico: Real And Nominal Rates Are Too High Mexico: Real And Nominal Rates Are Too High Despite widespread investor concerns, President AMLO has been running a very tight fiscal policy. At the end of 2019, the government had a primary surplus of 1% of GDP, and the overall deficit stood at 1.6%. In fact, given AMLO’s ideological approach to fiscal frugality, his government’s fiscal response to the COVID-19 pandemic to date has actually been less than what it can or should be. Similarly, monetary policy has been very tight. This is positive for creditors but negative for growth. The central bank has erred on the hawkish side and has a lot of room to reduce interest rates. Nominal and real interest rates in Mexico are among the highest in the EM universe (Chart I-1). Very tight fiscal policy means that monetary policy can be relaxed considerably. Interest rates in Mexico have a lot of downside.   Finally, the peso is reasonably cheap, according to the real effective exchange rate based on CPI and PPI measures (Chart I-2). Mexico’s macro risk premium relative to other EMs has been, in our opinion, wider than it should have been. On the other hand, there are considerable negatives, especially regarding the growth outlook: A year and a half into his mandate, president AMLO has not been able to secure the corporate sector’s confidence in his administration’s policies. The government was attempting to reverse this trend in the months leading up to the COVID-19 outbreak by announcing a public-private infrastructure package and improving relations with the US. Nevertheless, the decision to shun large corporations from the national fiscal response has once again weighed on business confidence. This will further reduce capital spending and hiring, prolonging the recession (Chart I-3). Chart I-2The Mexican Peso Is Cheap The Mexican Peso Is Cheap The Mexican Peso Is Cheap Chart I-3Business Confidence Plummets Again Business Confidence Plummets Again Business Confidence Plummets Again   The government’s fiscal response to the COVID-19 pandemic has been insufficient. The central government announced measures to increase funding for social and infrastructure programs and loans for households as well as small and medium businesses, amounting to a mere 3% of GDP. This is one of the lowest stimulus packages among major economies worldwide (Chart I-4). Chart I-4Mexico's Fiscal Response Is Poor Mexico, Korea & South Africa Mexico, Korea & South Africa Mexico is highly levered to the US economy. A deep contraction in American demand for consumer discretionary goods and international travel will suffocate Mexico’s export revenues. Exports of automobiles and tourism revenues together account for 37% of total goods and services exports, and 13% of GDP (Chart I-5). Balancing pros and cons, we recommend the following strategy for Mexican markets: Continue to overweight local currency bonds and sovereign credit within their respective EM benchmarks (Chart I-6). Orthodox fiscal and monetary policies warrant an overweight stance on fixed-income plays. Chart I-5Autos And Tourism Revenues Are Significant Autos And Tourism Revenues Are Significant Autos And Tourism Revenues Are Significant Chart I-6Mexico Versus EM: Domestic Bonds And Sovereign Credit Mexico Versus EM: Domestic Bonds And Sovereign Credit Mexico Versus EM: Domestic Bonds And Sovereign Credit   We reiterate our trade to receive Mexican 10-year swap rates. The only reason we are reluctant to be long cash domestic bonds is the potential for further currency depreciation. Finally, we are maintaining an overweight stance on equities, even though we acknowledge the very bad profit outlook. However, historically whenever Mexican interest rates have fallen relative to EM, Mexican stocks have typically outperformed the EM equity benchmark (Chart I-7). This is the primary rationale behind our equity overweight stance. Chart I-7Mexico vs. EM: Government Bond Yields Are Inversely Correlated To Stock Prices Mexico vs. EM: Government Bond Yields Are Inversely Correlated To Stock Prices Mexico vs. EM: Government Bond Yields Are Inversely Correlated To Stock Prices   Juan Egaña Research Associate juane@bcaresearch.com   South Korea: Bonds Offer Value Amid Looming Deflation The South Korean economy is facing strong deflationary pressures, requiring significant and additional rate cuts. Meanwhile, 10-year government bonds yield are still at 1.4%, 75 basis points over 10-year US Treasurys (Chart II-1). Hence, Korea’s bond yields offer good value for fixed-income investors and have considerable downside. We have been receiving 10-year swap rates in Korea since 2011 and are reiterating this recommendation: Chart II-2 shows that the GDP deflator has been negative since 2018, and core and trimmed mean consumer prices are flirting with deflation. Chart II-1Korean Government Bonds Yields: More Room To Fall Korean Government Bonds Yields: More Room To Fall Korean Government Bonds Yields: More Room To Fall Chart II-2The Korean Economy Is Flirting With Deflation The Korean Economy Is Flirting With Deflation The Korean Economy Is Flirting With Deflation   Falling prices amid elevated corporate and household debt levels – at 102% and 96% of GDP respectively – is toxic. The basis is price deflation increases real debt burdens. Notably, the debt service ratio for businesses and households is very high at 19.9% of GDP. There is no reason why Korea’s policy rate should not be reduced close to zero as is the case in advanced economies. Exports – which account for some 40% of GDP – are plunging. The business survey from Bank of Korea suggests exporters’ business sentiment plunged by a record in May and is close to 2008 levels, pointing to a dreadful export outlook. (Chart II-3) Domestic demand will remain weak, despite the large fiscal response to the COVID-19 outbreak. Business investment and hiring will be depressed for a while, undercutting consumer spending (Chart II-4). Chart II-3Exports In Freefall Exports In Freefall Exports In Freefall Chart II-4Less Investment Plan And Poor Employment Outlook Less Investment Plan And Poor Employment Outlook Less Investment Plan And Poor Employment Outlook Chart II-5Falling Residential Construction Permits Falling Residential Construction Permits Falling Residential Construction Permits Finally, residential investment was in the doldrums even before the COVID-19 outbreak. Chart II-5 illustrates that declining residential construction permits preclude lower residential construction for the rest of the year. The Bank of Korea will have to cut interest rates considerably this year. From a big-picture perspective, there is no reason why Korea’s policy rate should not be reduced close to zero as is the case in advanced economies. Korea’s economy shares many similarities with advanced economies like high debt levels and persistent deflationary pressures. On top of this, Korea is much more exposed to global trade, which makes its cyclical outlook worse, heralding substantial monetary easing. Exchange Rate Low interest rates could undermine the Korean won, even though the exchange rate has not historically been driven by interest rate differentials. The key driver of the won – shrinking global trade volumes and deflating tradable goods prices – warrants a cheaper currency to mitigate the negative impact on corporate profitability (Chart II-6). Chart II-6Deflating Export Prices Herald Currency Depreciation Deflating Export Prices Herald Currency Depreciation Deflating Export Prices Herald Currency Depreciation Chart II-7Deflating Semiconductor Prices... Deflating Semiconductor Prices... Deflating Semiconductor Prices...   Besides, deflation in DRAM prices (Chart II-7) as well as DRAM sales point to further currency depreciation and lower Korean tech stock prices (Chart II-8). Chart II-8...Does Not Bode Well For Tech Stocks Semiconductor Prices Are Still Deflating ...Does Not Bode Well For Tech Stocks Semiconductor Prices Are Still Deflating ...Does Not Bode Well For Tech Stocks Overall, a weak currency is needed to alleviate deflationary pressures currently present in the economy. Stocks We are negative on the KOSPI in absolute terms but continue to recommend that EM-dedicated equity portfolio investors overweight this bourse. Despite being a highly cyclical market, we believe the KOSPI’s outperformance will be due to its large weight in tech stocks. The latter will benefit from China’s ambitious tech-related infrastructure plan in the coming years. The plan includes construction of Information Transmission, Software and Information Technology Services, such as 5G networks, industrial internet and data centers. We expect total investment will reach between US$182 billion and $266 billion by the end of 2020, an increase of 30-50% over last year. Importantly, 40% of Korea’s semiconductor exports are purchased by China. We have been playing the semiconductor theme via Korea rather than Taiwan because the latter is a wild card amid escalating geopolitical tensions between the US and China. Our geopolitical team expects a flare up in US-China tensions ahead of US elections this year, and Taiwan could become one of the focal points. Bottom Line: Continue receiving 10-year swap rates, shorting the won against the US dollar and overweighting the KOSPI within an EM dedicated equity portfolio. Lin Xiang, CFA Research Analyst linx@bcaresearch.com South Africa: A Point Of No Return On Public Debt South Africa’s public debt is bound to surge to unsustainable levels: from 62% of GDP in 2019 to 95% of GDP by the end of 2021. If the government is forced to take over unsustainable debt from state-owned enterprises, which is very likely, it will push up the public debt-to-GDP ratio further by another nine percentage points to 104% of GDP. Table III-1 summarizes South Africa’s public debt projections using the following parameters and assumptions: To fight the COVID-19-induced economic crunch, President Cyril Ramaphosa recently announced a fiscal stimulus package of $26 billion (R500 billion), or 10% of GDP. Using recent government and central bank projections for 2020 and 2021, nominal GDP growth is expected to contract by 2.5% and expand 6.7%, respectively. Notably, fiscal revenue growth is expected to fall by 32% in nominal terms, according to recent comments by the Minister of Finance.1  Meanwhile, government spending will grow by 15%,2 and the primary fiscal deficit is expected to widen to 15.4% of GDP in 2020. Given that government forecasts often tend to be optimistic, chances are that both the primary deficit and public debt-to-GDP ratio will overshoot these forecasts. Finally, the sharp drop in domestic demand will increase the odds of a default among state-owned enterprises, with Eskom likely being a case in point. Current government guidelines require at least two thirds of Eskom’s R450 billion debt to be transferred to government balances in the event of default or anticipated default. In such a case, this increases the government debt-to-GDP ratio by an additional R350 billion, or 7% of GDP. Table III-1Projections For South Africa Fiscal Position And Public Debt Mexico, Korea & South Africa Mexico, Korea & South Africa Altogether, the public debt-to-GDP ratio will surge to 104% of GDP by the end of 2021 (Chart III-1). With public debt above 100% of GDP, interest rates well above nominal GDP and the government running large primary deficits, debt dynamics will become unsustainable. To avoid a public debt crisis, the government should either run large primary surpluses, which is unfeasible anytime soon, or bring down government borrowing costs to push up nominal GDP above interest rates (Chart III-2). Chart III-1Public Debt-To-GDP Will Balloon To 104%! Public Debt-To-GDP Will Balloon To 104%! Public Debt-To-GDP Will Balloon To 104%! Chart III-2Unsustainable Gap Between Local Yields And Nominal Growth Unsustainable Gap Between Local Yields And Nominal Growth Unsustainable Gap Between Local Yields And Nominal Growth   The latter option is the only one that is politically feasible. But to do so, the central bank needs to resort to the monetization of public debt. The central bank (SARB) has already taken the first step to bring down bond yields by buying government bonds in the secondary market. While the rationale of that was to cover foreign investors’ selling of local currency bonds, it amounts to nothing else but quantitative easing, or public debt monetization. Ultimately, the outcome of large fiscal deficits and public debt monetization is a weaker currency. As such, debt monetization is a fait accompli in South Africa. Monetizing part of the government’s debt will help reduce real borrowing costs and at the same time reflate nominal GDP growth, thereby boosting government revenues. Ultimately, the outcome of large fiscal deficits and public debt monetization is a weaker currency. If foreigners continue to sell the local currency bond market, the SARB and commercial banks will need to buy more government debt, creating even more money. This is why we expect the rand to continue depreciating. Investment Recommendations Chart III-3The Rand Could Drop Further Given Public Debt Dynamics The Rand Could Drop Further Given Public Debt Dynamics The Rand Could Drop Further Given Public Debt Dynamics The currency will likely get cheaper provided the rising odds of outright public debt monetization (Chart III-3). Continue shorting the rand versus the US dollar. We are initiating a new position of receiving 2-year swap rates. Odds are that the central bank will cut rates further in the months to come. Remain underweight local currency bonds in an EM-dedicated portfolio. Even though local domestic rates will likely fall, South African bonds will not outperform the EM benchmark on a total return in US dollar basis, mostly due to chronic currency depreciation. Finally, investors should underweight sovereign credit (government US dollar bonds) due to the unsustainable public debt dynamics. Dedicated EM equity portfolio investors should maintain a below-benchmark allocation to this bourse. Andrija Vesic Associate Editor andrijav@bcaresearch.com   Footnotes 1     The Minister of Finance made remarks about tax revenue falling by 32% in nominal terms. Tax revenues represent almost 100% of overall revenue. 2     Overall fiscal package is estimated to be 3% of GDP. This excludes reprioritization in 2020 around R130 billion & loan guarantee scheme of R200 billion. Overall total additional spending amounts to R170 billion in 2020 fiscal year. Equities Recommendations Currencies, Credit And Fixed-Income Recommendations

Global semiconductor share prices have continued to hit new highs, even though there has not been any recovery (positive growth) in global semiconductor sales or in their corporate earnings (EPS). Global semiconductor sales bottomed on a rate-of-change basis in June, but their annual growth rate was still negative in December. In the meantime, global semi share prices have been rallying since January 2019. This divergence between stock prices and revenue of global semiconductor stocks is unprecedented (Chart II-1). Chart II-1Global Semiconductor Market: Sales & Share Prices Over-Hyped Global Semi Share Prices Global Semiconductor Market: Sales & Share Prices Over-Hyped Global Semi Share Prices Global Semiconductor Market: Sales & Share Prices Over-Hyped Global Semi Share Prices Odds are that global semi stocks in general, and Asian ones in particular, will experience a pullback in the coming weeks. The coronavirus outbreak will likely dampen expectations related to the speed of 5G adoption and penetration in China. Critically, China accounted for 35% of global semiconductor sales in 2019, versus 19% for the US and 10% for the whole of Europe. In brief, semiconductor demand from China is now greater than the US and European demand combined. Furthermore, the latest news that the US administration is considering changing its regulations to prevent shipments of semiconductor chips to China’s Huawei Technologies from global companies - including Taiwan's TSMC - could hurt chip stocks further. Since Huawei Technologies is the global leader in 5G networks and smartphones, the ban, if implemented, will instigate a sizable setback to 5G adoption in China and elsewhere. Our updated estimate of global 5G smartphone shipments is between 160 million and 180 million units in 2020, which is below the median of industry expectations of 210 million units (Table II-1). The key reasons why the industry’s expectations are unreasonably high, in our opinion, are as follows: Chinese demand for new smartphones will likely stay weak (Chart II-2). The mainland smartphone market has become extremely saturated, with 1.3 billion units having been sold in just the past three years – nearly equaling the entire Chinese population. Table II-1Industry Forecasts Of The 2020 Global 5G- Smartphone Shipments EM: Growing Risk Of A Breakdown EM: Growing Risk Of A Breakdown Chart II-2Chinese Smartphone Demand: Further Decline In 2020 Chinese Smartphone Demand: Further Decline In 2020 Chinese Smartphone Demand: Further Decline In 2020     Chinese official data show that each Chinese household owned 2.5 phones on average in 2018, and that the average household size was about three persons (Chart II-3). This suggests that going forward nearly all potential phone demand in China is for replacement phones, and that there is no urgent need for households to buy new phones. Chart II-3Chinese Households: No Urgent Need For A New Phone Chinese Households: No Urgent Need For A New Phone Chinese Households: No Urgent Need For A New Phone The Chinese government’s boost to 5G infrastructure investment will likely increase annual installed 5G base stations from 130,000 units last year to about 600,000 to 800,000 this year. However, the total number of 5G base stations will still only account for about 7-9% of total base stations in China in 2020. Hence, geographical coverage will not be sufficiently wide enough to warrant a very high rate of 5G smartphone adoption and penetration. From Chinese consumers’ perspectives, a 5G phone in 2020 will be a ‘nice-to-have,’ but not a ‘must-have.’ Given increasing economic uncertainty and many concerns related to the use of 5G phones, mainland consumers may delay their purchases into 2021 when 5G phone networks will have more geographic coverage.  The number of 5G phone models on the market is expanding, but not that quickly. Consumers may take their time to wait for more models to hit the market before making a 5G phone purchase. For example, Apple will release four 5G phone models, but only in September 2020. Moreover, the price competition between 5G and 4G phones is getting increasingly intense. Smartphone producers have already started to cut prices of their 4G phones aggressively. For example, the price of Apple’s iPhone XS, released in September 2018, has already dropped by about 50% in China. Outside of China, 5G infrastructure development will be much slower. The majority of developed countries will likely give in to pressure from the US and limit their use of Huawei 5G equipment. This will delay infrastructure installation and adoption of 5G throughout the rest of the world because Huawei has the leading and cheapest 5G technology. In 2019, China accounted for about 70% of worldwide 5G smartphone shipments. We reckon that in 2020 Chinese 5G smartphone shipments will be between 120 million and 130 million units. Assuming this accounts for about 70-75% of the world shipment of 5G phones this year, we arrive at our estimate of global 5G smartphone shipments of between 160 million and 180 million units. Overall, investors are pricing global semi stocks using the pace and trajectory of 4G smartphones adoption. However, in 2020 the number and speed of 5G phone penetration will continue lagging that of 4G ones when the latter were introduced in December 2013 (Chart II-4). We agree that 5G technology is revolutionary, and its adoption and penetration will surge in the coming years. Nevertheless, we still believe global semi share prices are presently overhyped by unreasonably optimistic 2020 projections (Chart II-5).  Chart II-4China 5G-Adoption Pace: Slower Than The Case With 4G China 5G-Adoption Pace: Slower Than The Case With 4G China 5G-Adoption Pace: Slower Than The Case With 4G Chart II-5Net Earnings Of Global Semi Sector: Too Optimistic? Net Earnings Of Global Semi Sector: Too Optimistic? Net Earnings Of Global Semi Sector: Too Optimistic?   Investment Implications Global semi stocks’ valuations are very elevated, as shown in Chart II-6 and Chart II-7. Besides, semi stocks are overbought, suggesting they could correct meaningfully if lofty growth expectations currently baked into their prices do not materialize in the first half of this year. Chart II-6Global Semi Stocks Valuations: Very Elevated Global Semi Stocks Valuations: Very Elevated Global Semi Stocks Valuations: Very Elevated Chart II-7Global Semi Stocks’ Valuations: Very Elevated Global Semi Stocks Valuations: Very Elevated Global Semi Stocks Valuations: Very Elevated   The coronavirus outbreak and the resulting delay in 5G phone sales in China in the first half of 2020, along with US pressure on global semi producers not to sell to Huawei, will likely trigger a pullback in semiconductor equities. We recommend patiently waiting for a better entry point for absolute return investors. Within the EM equity universe, we have not been underweight Asian semi stocks because of our negative outlook for the overall EM equity benchmark. We remain neutral on Taiwan and overweight Korea. The reason is that DRAM makers such as Samsung and Hynix have rallied much less than TSMC. Besides, geopolitical risks in relation to Taiwan in general and TSMC in particular are rising, warranting a more defensive stance on Taiwanese stocks relative to Korean equities. Ellen JingYuan He Associate Vice President ellenj@bcaresearch.com

Dear Client, Over the past two weeks, I have been in Asia visiting BCA’s clients. Next week’s Report, on November 20 will be a recap of my observations from the road. This week we are sending you a Special Report on global semiconductor stock performance published by our Emerging Markets Strategy service, authored by my colleague Ellen JingYuan He.  This Special Report offers great insights on the development of 5G network industry, global demand beyond 5G smartphones, as well as investment implications derived from the research. I hope you find it interesting and insightful. Best regards, Jing Sima, China Strategist   Highlights Since early this year, global semiconductor stock prices have been front-running a demand recovery that has not yet begun. There is strong industry optimism surrounding a potential demand boost for semiconductors from the rollout of 5G networks and phones in 2020. Yet we expect actual 2020 Chinese 5G smartphone shipments to fall considerably short of what industry observers expect, especially in the first half of the year. Global semiconductor stocks are over-hyped. Even though momentum could push them higher in the short term, we believe there will be a better entry point in the coming months. Given that Korean semiconductor stocks have lagged, we are upgrading Korean tech stocks and the KOSPI to overweight within the EM equity benchmark. Feature Global semiconductor stock prices have been rallying strongly, increasingly diverging from global semiconductor sales since early January. The former have risen to new highs, while the latter have remained in deep contraction (Chart 1). Chart 1A Puzzle: Semiconductors Stock Prices Skyrocketed When Sales Remain In A Deep Contraction Global Semiconductor Market: Sales & Share Prices A Puzzle: Semiconductors Stock Prices Skyrocketed When Sales Remain In A Deep Contraction Global Semiconductor Market: Sales & Share Prices A Puzzle: Semiconductors Stock Prices Skyrocketed When Sales Remain In A Deep Contraction We are puzzled by such a dramatic divergence between share prices and the industry’s top line. After all, the ongoing contraction in worldwide semiconductor sales has been broad-based across both regions and the majority of top 10 semiconductor companies (Charts 2 and 3). Chart 2A Broad-Based Contraction Across All Regions… A Broad-Based Contraction Across All Regions... A Broad-Based Contraction Across All Regions... Chart 3…And Most Top Semiconductor Companies ...And Most Top Semiconductor Companies ...And Most Top Semiconductor Companies   In our June1 report, we argued that world semiconductor sales would continue to shrink through the remainder of 2019. This view has played out, but global semiconductor share prices have surged and outperformed the global equity benchmark.  Global semiconductor stock prices have been front-running a demand recovery that has not yet begun. It seems the market has been looking beyond the current weakness. It currently expects a potential demand boost for semiconductors from 5G phones in 2020 on the back of rising hopes of a US-China trade conflict resolution. Is such hype about 5G network and corresponding shipments justified? Our research leads us to contend that global semiconductor sales will likely post only low- to middle-single-digit growth in 2020, with most of the recovery back loaded in the second half of the year. Hype over 5G phones among industry participants and investors may continue pushing semiconductor share prices higher in the near term. However, the odds are that the reality of tepid semiconductor sales growth will likely set in early next year, and semiconductor stocks will correct considerably. In short, we do not recommend chasing the rally. There will be a better entry point in the months ahead. 5G-Smartphones: The Savior Of Semiconductor Demand? Chart 4Semiconductor Sales Are Still Contracting At A Double-Digit Rate Strong Global 5G-Smartphone Shipments In 2020? Strong Global 5G-Smartphone Shipments In 2020? The primary driver behind the rally in semiconductor share prices is strong optimism among major semiconductor producers and investors about a rapid ramp-up of global 5G-smartphone adoption. In addition, the market is also holding onto a good amount of hope for a US-China trade conflict resolution, which will also facilitate the pace of global 5G deployment. Mobile phones account for the largest share (29%) of global semiconductor revenue. The industry expects strong global 5G-smartphone shipments in 2020 to spur a meaningful recovery in semiconductor demand (Chart 4). Table 1 shows a list of estimates for 2020 global 5G-smartphone shipments by major semiconductor companies, industry analysts and investors, ranging from 120 million to 225 million units, with a mean of 180 million units. Table 1Market Forecasts Of In 2020 Global 5G-Smartphone Shipments Semiconductor Stocks: Is The 5G Hype Warranted? Semiconductor Stocks: Is The 5G Hype Warranted? In particular, Taiwan Semiconductor Manufacturing Company (TSMC), the world’s largest dedicated integrated circuit (IC) foundry, recently almost doubled its forecast for 5G smartphone penetration for 2020 to a mid-teen percentage from a single-digit percentage forecast made just six months ago. Given that global smartphone shipments currently stand at roughly 1.4 billion units per year, a 15% penetration rate would translate into 210 million units of 5G smartphone shipments in 2020. Meanwhile, Qualcomm, the world's largest maker of mobile application processors and baseband modems, last week predicted that 2020 global 5G smartphone shipments will range between 175 million units and 225 million units. We agree that 5G smartphone sales in 2020 will increase sharply from currently very low levels, but we also believe the penetration pace estimated by the industry is optimistic. The basis for our conclusion is as follows: Chart 5So Far, China 5G-Adoption Pace Has Been Much Slower Than Its 4G So Far, China 5G-Adoption Pace Has Been Much Slower Than Its 4G So Far, China 5G-Adoption Pace Has Been Much Slower Than Its 4G 5G-smartphone shipments in China will largely determine the pace of worldwide 5G-phone shipments. The country will be the world leader in the 5G smartphone market due to the government’s promotion of it and the advanced 5G technology held by China's largest telecom equipment producer, Huawei. China announced the debut of the 5G-era on June 6. Since then, total 5G-smartphone shipments have been only about 800,000 units through the end of September. In terms of the pace of penetration (5G-smartphone shipments as a share of total mobile phone shipments during the first three months of launch), the rate was a mere 0.3%. In comparison with the debut of the 4G-era in December 2013, shipments of 4G phones in China were significantly larger, and their adoption rate was much faster (Chart 5). During the first three months of the 4G launch, 4G phone shipments were 9.7 million units, reaching 10% of total smartphone shipments. Here are the most important reasons behind what will be a much slower penetration pace for 5G smartphones in China compared with the 4G rollout. We agree that 5G smartphone sales in 2020 will increase sharply from currently very low levels, but we also believe the penetration pace estimated by the industry is optimistic. Market saturation: The Chinese smartphone market has become much more saturated than it was six years ago when 4G was launched. Since then, there have been about 2.3 billion units of 4G smartphones sold, with 1.3 billion units sold in the past three years – nearly equaling the total Chinese population. This means the replacement need in China is low. High prices: 5G smartphones in China are currently much more expensive than 4G ones. 5G phone prices range from RMB 4000-7000 in China, while most of the 4G ones sell within the range of RMB 1000-3000. According to data from QuestMobile, a professional big data intelligence service provider in China's mobile internet market, in the first half of 2019, about 41% of smartphones were sold at RMB 1000-2000, about 30% at RMB 2000-3000, and only 10% at RMB 4000 and above. Functionality: At the moment, except for faster data download/upload speed, 5G smartphones do not offer much more functionality than 4G ones. Back in 2014, 4G phones had much more attractive features than 3G. For example, while 3G smartphones only allowed audio and picture transmission, those with 4G enabled video chatting and high-quality streaming video. In addition, for now, there are very few smartphone apps that can only be used for 5G phones. 5G Infrastructure: Presently, there is only very limited geographical coverage of 5G base stations. The number of 5G base stations is estimated to be 130 thousand units this year, only accounting for 1.6% of total base stations in China. In comparison, 65% of all Chinese base stations are 4G-enabled.  Meanwhile, to cover the same region, the number of 5G base stations needs to at least double that of 4G ones. It will take at a minimum two or three years to develop decent coverage of 5G base stations. Besides, the cost of building 5G-enabled infrastructure is much more expensive than the deployment of the 4G ones. There are two types of 5G networks: Non-standalone (NSA) and Standalone (SA). The 5G data transmission speed is significantly faster in SA mode than in NSA mode. However, the deployment cost of the SA network is much higher than the cost for NSA networks, as the latter can be built from existing 4G networks, but the former cannot. Critically, the Chinese government recently announced only SA-compatible 5G smartphones will be allowed to have access to the 5G network in China, starting January 1, 2020. This signals that the focus of future 5G network development will be centered around SA mode instead of this year’s NSA mode. Over 90% of China’s 5G network was NSA mode in 2019. Building a 5G SA network will take longer and cost more.  The market expects China to build as much as 1 million units of 5G base stations in 2020. Even if this goal is achieved, it only accounts for about 11% of total Chinese base stations. Chart 6Chinese Smartphone Sales: Still In Contraction Chinese Smartphone Sales: Still In Contraction Chinese Smartphone Sales: Still In Contraction Lack of variety of SA-compatible 5G-phone models. There are also limited options for SA-compatible 5G smartphones models. So far, even though Huawei, Xiaomi, Vivo, Oppo, ZTE and Samsung have all released 5G smartphones, only models from Huawei work under SA networks.2 All others only work under the NSA network. Hence, the variety of SA-compatible 5G phone models is very limited. This will likely delay sales of 5G phones in China. Many more models of SA-compatible 5G smartphones will likely be released only in the second half of next year, which may both drive down 5G smartphone prices and attract more buyers. Consumer spending slowdown: 4G smartphones can meet the needs of the majority of users, and most users have purchased a new phone within the past three years. With elevated economic uncertainty and slowing income growth, a larger proportion of people in China may decide to delay upgrading from 4G-phones to much more expensive 5G ones. This echoes a continuing decline in Chinese smartphone sales (Chart 6). Overall, from Chinese consumers’ perspective, a 5G phone in 2020 will be a nice-to-have, but not a must-have. Given all the aforementioned factors, our best guess for 2020 Chinese 5G smartphone shipments is 40-60 million units, with a larger proportion occurring in the second half of the year.  From Chinese consumers’ perspective, a 5G phone in 2020 will be a nice-to-have, but not a must-have. As China is much more aggressive in moving to 5G network adaptation than other large economies, we share industry experts’ forecasts that China will account for 50% of total global 5G shipments. Provided our estimate for China is about 50 million units, our global forecast for 5G phone shipments in 2020 comes to about 100 million units worldwide. This is substantially lower than industry and analyst average estimates of 180 million units (see Table 1 on page 4). Notably, rising 5G smartphone sales will cannibalize some 4G-phone demand. Consequently, aggregate demand for semiconductors will not grow, but the share of high-valued-added chips in the overall product mix will rise. Bottom Line: The penetration pace of 5G smartphones will be meaningfully slower than both the semiconductor producers and analysts expect. Most likely, a meaningful recovery in global aggregate smartphone sales will not occur over the next six months. We suspect the positive impact of 5G phone sales will be felt by global semiconductor producers largely in the second half of 2020. Semiconductor Demand Beyond 5G In terms of end usage, except smartphones, the top five end uses for semiconductors are personal computers (PCs) (12%), servers (11%), diverse consumer products (12%), automotive (10%), and industrial electronics (9%). Structural PC demand is down, but sales have been more or less flat in the past three years (Chart 7). Next year, commercial demand may accelerate as enterprises work through the remainder of their Windows 10 migration. However, household demand is still facing strong competition from tablets. Overall, we expect PC demand to remain stagnant. Global server shipments sank deeper into contraction in the second quarter of this year due to a slowdown in purchasing from cloud providers and hyperscale customers. They may stay in moderate contraction over the next six months as global economic uncertainty remain elevated, which may discourage enterprises’ investment plans (Chart 8). Chart 7Structural PC Demand Is Stagnant And Will Remain So In 2020 Global PCs Sales: Deeply Saturated Structural PC Demand: Remain Stagnant In 2020 Global PCs Sales: Deeply Saturated Structural PC Demand: Remain Stagnant In 2020 Chart 8Global Server Shipments: A Moderate Contraction In 2020 Global Server Shipment: Are In Contraction Global Server Shipments: Moderate Contraction In 2020 Global Server Shipment: Are In Contraction Global Server Shipments: Moderate Contraction In 2020 Chart 9Automotive-Related Semiconductor Demand: A Moderate Growth Ahead Automotive-Related Semiconductor Demand: A Moderate Growth Ahead Automotive-Related Semiconductor Demand: A Moderate Growth Ahead Chinese auto sales – about 30% of the world total – will likely stage a rate-of-change improvement, moving from deep to mild contraction or stagnation over the next six months.3 Increasing penetration of new energy vehicles and continuing 5G deployment may still result in moderate growth in auto-related semiconductor demand (Chart 9). Semiconductor demand from diverse consumer products slightly declined in the third quarter, with robust growth in tablets, eReaders and portable navigation devices, and contraction in all other subsectors including TV sets, gaming, printers and images, cameras and set-top boxes (Chart 10). This may remain in slight contraction or stagnation over the next three to six months. Automation and 5G deployment will likely continue to increase semiconductor sales in the industrial sector (Chart 11).  Chart 10Semiconductor Demand From Consumer Products: A Slight Contraction Or Stagnation Ahead Semiconductor Stocks: Is The 5G Hype Warranted? Semiconductor Stocks: Is The 5G Hype Warranted? Chart 11Industrial Semiconductor Demand: More Upside Ahead Industrial Semiconductor Demand: More Upside Ahead Industrial Semiconductor Demand: More Upside Ahead   Chart 12Memory Prices Still Signal Sluggish Semiconductor Demand Memory Prices Still Signal Sluggish Semiconductor Demand Memory Prices Still Signal Sluggish Semiconductor Demand Overall, demand recovery has not yet begun. The lack of price recovery in DRAM prices after 18 months of declines and still-low NAND prices are also signaling sluggish semiconductor demand (Chart 12). Bottom Line: Odds are that global semiconductor demand in sectors other than smartphones will show improvement in terms of rate of change, but will still likely be flat in 2020. TSMC Sales: A Harbinger Of Industry Recovery? TSMC, the world’s biggest semiconductor company, posted a revival in sales over four consecutive months from June to September. Do TSMC sales lead global semiconductor sales? The answer is not always. TSMC sales do not always correlate well with global semiconductor sales (Chart 13). For example, TSMC sales diverged from global semiconductor sales in 2017-‘18 and 2013-‘14. So what are the reasons for strong increase in TSMC sales? First, it reflects market share rotation in the global smartphone market in favor of smartphone producers that use TSMC-fabricated chips. Chart 13TSMC Sales Do Not Always Lead Global Semiconductor Sales TSMC Sales Do Not Always Lead Global Semiconductor Sales TSMC Sales Do Not Always Lead Global Semiconductor Sales Demand from the global smartphone sector contributes to almost half of TSMC’s total revenue. Apple and Huawei are TSMC’s two top customers. The most recent report from market research firm Canalys shows that while Apple’s smartphone shipments declined 7% year-on-year last quarter, Huawei’s shipments soared 29%.4 Combined, smartphone shipments from these two companies still jumped nearly 12% year-on-year in the third quarter of the year. This has increased their market share in the global smartphone market to 31% now from 28% a year ago. Second, rising TSMC sales also reflect market share rotation in the global server market, in particular rising shipments and growing market share of servers using AMD high-performing-computing (HPC) chips instead of Intel ones. AMD’s 7nm Epyc CPU, launched this August and manufactured by TSMC, has been taking share from Intel in the global server market. This has driven the increase in TSMC’s revenue from the HPC sector.  Third, the share of value-added products (high-end chips) in TSMC’s product mix has been rising rapidly. TSMC’s share of revenue from 7nm technology jumped from 21% to 27% in the third quarter, as most of Apple’s and Huawei’s chips and all of AMD’s Epyc CPUs are 7nm-based. Back in the third quarter of 2018, TSMC’s 7nm business only accounted for 11% of its total revenue. Chart 14Both TSMC Sales And Taiwanese PMI Could Continue To Improve While Global Semiconductor Sales Remain In Contraction Both TSMC Sales And Taiwanese PMI Could Continue To Improve While Global Semiconductor Sales Remain In Contraction Both TSMC Sales And Taiwanese PMI Could Continue To Improve While Global Semiconductor Sales Remain In Contraction Finally, although internet of things (IoT) and automotive chips only account for 9% and 4% of TSMC’s total share of revenue respectively, strong growth in both segments –33% year-on-year in IoT and 20% year-on-year in automotive – indeed shows exceptional demand in these two sectors in a weakening global economic environment. As IoT and automotive development will highly rely on global 5G infrastructure development, their impact will be meaningful once the global 5G network becomes well advanced and widely installed. To conclude, while a 40% boost in TSMC’s capital spending indeed paints a positive picture on global semiconductor demand over the longer term, rising TSMC sales do not mean an imminent and strong recovery in the global semiconductor sector is in the works. Huawei is the global 5G technology leader and the major supplier in both 5G-network equipment and 5G smartphones; the company will be a major revenue contributor to TSMC. As Huawei will likely place more orders to TSMC for chip fabrication, this will likely result in further improvement in TSMC’s sales (Chart 14). Bottom Line: Rising TSMC sales do not necessarily herald an imminent and robust cyclical recovery in the global semiconductor sector.  Investment Conclusions Global semiconductor stock prices have been front running a recovery that has not yet begun. In addition, there is still uncertainty about the technology aspect of US-China trade negotiations. The US will likely continue to have Huawei and other Chinese high-tech companies on its trade-ban list – its so-called Entity List. TSMC sales do not always correlate well with global semiconductor sales. Notably, global semiconductor sales and profits are still in deep contraction, while share prices are at all-time highs (Chart 15). As a result, semiconductor stocks’ multiples have spiked to their previous highs (Chart 16). Chart 15Semiconductor Companies Profits: Still In Deep Contraction Semiconductor Companies Profits: Still In Deep Contraction Semiconductor Companies Profits: Still In Deep Contraction Chart 16Elevated Semiconductor Stocks Multiples Elevated Semiconductor Stocks Multiples Elevated Semiconductor Stocks Multiples     While it is common for share prices to rally ahead of a business cycle/profit revival, we believe a true recovery will only emerge in spring 2020, and it will initially be much more subdued than industry watchers and investors expect. In the near term, strong momentum could still push semiconductor stock prices higher. However, the reality will then set in and there will be an air pocket before a more sustainable bull market emerges.   Our US Equity Investment Strategy earlier this week downgraded S&P semiconductor equipment companies to underweight and put the S&P Semiconductors Index on a downgrade alert.5 Their newly created top-down semiconductor profit growth model warns that an earnings recovery is not yet imminent (Chart 17). For EM-dedicated equity managers, we have been neutral on Asian semiconductor sectors. We continue to recommend a market-weight allocation to Taiwan’s overall market, while we are upgrading the Korean technology sector from a neutral allocation to overweight. Korean semiconductor stocks have rallied much less than their global peers. Hence, the risk of a major relapse is lower. Given that we have been overweight non-tech Korean stocks, upgrading tech stocks to overweight means we will be overweight the KOSPI within the EM equity benchmark (Chart 18). Chart 17Semiconductor Earnings Recovery: Not Imminent Semiconductor Earnings Recovery: Not Imminent Semiconductor Earnings Recovery: Not Imminent Chart 18Upgrade Korean Tech Stocks And Overweight KOSPI Within EM Upgrade Korean Tech Stocks And Overweight KOSPI Within EM Upgrade Korean Tech Stocks And Overweight KOSPI Within EM   Meanwhile, we remain long the Bloomberg Asia-Pacific Semiconductor Index and short the S&P 500 Semiconductor Index. This trade has produced a 7% gain since its initiation on June 13, 2019. The Bloomberg Asia-Pacific Semiconductor index has 12 stocks. Samsung and TSMC account for 38% and 37% of the index, respectively. The S&P 500 Semiconductor Index has 13 stocks. Intel, Broadcom, Texas Instruments and Qualcomm are the top five constituents, together accounting for nearly 77% of the index. Although the US and China may reach a temporary trade deal, the US will continue to restrict sales of tech products and high-end semiconductors to China. As a result, these US semiconductor companies, most of which are IC designing companies, will likely experience a more subdued than expected recovery in sales. Ellen JingYuan He Associate Vice President ellenj@bcaresearch.com Footnotes   1 Please see Emerging Markets Strategy Special Report "The Global Semiconductor Sector: Is A Cyclical Upturn Imminent?" dated June 13, 2019, available at ems.bcaresearch.com 2 https://www.guancha.cn/ChanJing/2019_09_21_518748.shtml http://www.cac.gov.cn/2019-10/23/c_1573361796389322.htm 3 Please see Emerging Markets Strategy Special Report "Chinese Auto Demand: Time For A  Recovery?" dated October 17, 2019, available at ems.bcaresearch.com 4 https://www.canalys.com/analysis/smartphone+analysis 5 Please see US Equity Strategy Special Report "Defying Gravity," dated November 4, 2019, available at uses.bcaresearch.com   Cyclical Investment Stance Equity Sector Recommendations

Highlights Since early this year, global semiconductor stock prices have been front-running a demand recovery that has not yet begun. There is strong industry optimism surrounding a potential demand boost for semiconductors from the rollout of 5G networks and phones in 2020. Yet we expect actual 2020 Chinese 5G smartphone shipments to fall considerably short of what industry observers expect, especially in the first half of the year. Global semiconductor stocks are over-hyped. Even though momentum could push them higher in the short term, we believe there will be a better entry point in the coming months. Given that Korean semiconductor stocks have lagged, we are upgrading Korean tech stocks and the KOSPI to overweight within the EM equity benchmark. Feature Global semiconductor stock prices have been rallying strongly, increasingly diverging from global semiconductor sales since early January. The former have risen to new highs, while the latter have remained in deep contraction (Chart 1). Chart 1A Puzzle: Semiconductors Stock Prices Skyrocketed When Sales Remain In A Deep Contraction Global Semiconductor Market: Sales & Share Prices A Puzzle: Semiconductors Stock Prices Skyrocketed When Sales Remain In A Deep Contraction Global Semiconductor Market: Sales & Share Prices A Puzzle: Semiconductors Stock Prices Skyrocketed When Sales Remain In A Deep Contraction We are puzzled by such a dramatic divergence between share prices and the industry’s top line. After all, the ongoing contraction in worldwide semiconductor sales has been broad-based across both regions and the majority of top 10 semiconductor companies (Charts 2 and 3). Chart 2A Broad-Based Contraction Across All Regions… A Broad-Based Contraction Across All Regions... A Broad-Based Contraction Across All Regions... Chart 3…And Most Top Semiconductor Companies ...And Most Top Semiconductor Companies ...And Most Top Semiconductor Companies   In our June1 report, we argued that world semiconductor sales would continue to shrink through the remainder of 2019. This view has played out, but global semiconductor share prices have surged and outperformed the global equity benchmark.  Global semiconductor stock prices have been front-running a demand recovery that has not yet begun. It seems the market has been looking beyond the current weakness. It currently expects a potential demand boost for semiconductors from 5G phones in 2020 on the back of rising hopes of a US-China trade conflict resolution. Is such hype about 5G network and corresponding shipments justified? Our research leads us to contend that global semiconductor sales will likely post only low- to middle-single-digit growth in 2020, with most of the recovery back loaded in the second half of the year. Hype over 5G phones among industry participants and investors may continue pushing semiconductor share prices higher in the near term. However, the odds are that the reality of tepid semiconductor sales growth will likely set in early next year, and semiconductor stocks will correct considerably. In short, we do not recommend chasing the rally. There will be a better entry point in the months ahead. 5G-Smartphones: The Savior Of Semiconductor Demand? Chart 4Semiconductor Sales Are Still Contracting At A Double-Digit Rate Strong Global 5G-Smartphone Shipments In 2020? Strong Global 5G-Smartphone Shipments In 2020? The primary driver behind the rally in semiconductor share prices is strong optimism among major semiconductor producers and investors about a rapid ramp-up of global 5G-smartphone adoption. In addition, the market is also holding onto a good amount of hope for a US-China trade conflict resolution, which will also facilitate the pace of global 5G deployment. Mobile phones account for the largest share (29%) of global semiconductor revenue. The industry expects strong global 5G-smartphone shipments in 2020 to spur a meaningful recovery in semiconductor demand (Chart 4). Table 1 shows a list of estimates for 2020 global 5G-smartphone shipments by major semiconductor companies, industry analysts and investors, ranging from 120 million to 225 million units, with a mean of 180 million units. Table 1Market Forecasts Of In 2020 Global 5G-Smartphone Shipments Semiconductor Stocks: Is The 5G Hype Warranted? Semiconductor Stocks: Is The 5G Hype Warranted? In particular, Taiwan Semiconductor Manufacturing Company (TSMC), the world’s largest dedicated integrated circuit (IC) foundry, recently almost doubled its forecast for 5G smartphone penetration for 2020 to a mid-teen percentage from a single-digit percentage forecast made just six months ago. Given that global smartphone shipments currently stand at roughly 1.4 billion units per year, a 15% penetration rate would translate into 210 million units of 5G smartphone shipments in 2020. Meanwhile, Qualcomm, the world's largest maker of mobile application processors and baseband modems, last week predicted that 2020 global 5G smartphone shipments will range between 175 million units and 225 million units. We agree that 5G smartphone sales in 2020 will increase sharply from currently very low levels, but we also believe the penetration pace estimated by the industry is optimistic. The basis for our conclusion is as follows: Chart 5So Far, China 5G-Adoption Pace Has Been Much Slower Than Its 4G So Far, China 5G-Adoption Pace Has Been Much Slower Than Its 4G So Far, China 5G-Adoption Pace Has Been Much Slower Than Its 4G 5G-smartphone shipments in China will largely determine the pace of worldwide 5G-phone shipments. The country will be the world leader in the 5G smartphone market due to the government’s promotion of it and the advanced 5G technology held by China's largest telecom equipment producer, Huawei. China announced the debut of the 5G-era on June 6. Since then, total 5G-smartphone shipments have been only about 800,000 units through the end of September. In terms of the pace of penetration (5G-smartphone shipments as a share of total mobile phone shipments during the first three months of launch), the rate was a mere 0.3%. In comparison with the debut of the 4G-era in December 2013, shipments of 4G phones in China were significantly larger, and their adoption rate was much faster (Chart 5). During the first three months of the 4G launch, 4G phone shipments were 9.7 million units, reaching 10% of total smartphone shipments. Here are the most important reasons behind what will be a much slower penetration pace for 5G smartphones in China compared with the 4G rollout. We agree that 5G smartphone sales in 2020 will increase sharply from currently very low levels, but we also believe the penetration pace estimated by the industry is optimistic. Market saturation: The Chinese smartphone market has become much more saturated than it was six years ago when 4G was launched. Since then, there have been about 2.3 billion units of 4G smartphones sold, with 1.3 billion units sold in the past three years – nearly equaling the total Chinese population. This means the replacement need in China is low. High prices: 5G smartphones in China are currently much more expensive than 4G ones. 5G phone prices range from RMB 4000-7000 in China, while most of the 4G ones sell within the range of RMB 1000-3000. According to data from QuestMobile, a professional big data intelligence service provider in China's mobile internet market, in the first half of 2019, about 41% of smartphones were sold at RMB 1000-2000, about 30% at RMB 2000-3000, and only 10% at RMB 4000 and above. Functionality: At the moment, except for faster data download/upload speed, 5G smartphones do not offer much more functionality than 4G ones. Back in 2014, 4G phones had much more attractive features than 3G. For example, while 3G smartphones only allowed audio and picture transmission, those with 4G enabled video chatting and high-quality streaming video. In addition, for now, there are very few smartphone apps that can only be used for 5G phones. 5G Infrastructure: Presently, there is only very limited geographical coverage of 5G base stations. The number of 5G base stations is estimated to be 130 thousand units this year, only accounting for 1.6% of total base stations in China. In comparison, 65% of all Chinese base stations are 4G-enabled.  Meanwhile, to cover the same region, the number of 5G base stations needs to at least double that of 4G ones. It will take at a minimum two or three years to develop decent coverage of 5G base stations. Besides, the cost of building 5G-enabled infrastructure is much more expensive than the deployment of the 4G ones. There are two types of 5G networks: Non-standalone (NSA) and Standalone (SA). The 5G data transmission speed is significantly faster in SA mode than in NSA mode. However, the deployment cost of the SA network is much higher than the cost for NSA networks, as the latter can be built from existing 4G networks, but the former cannot. Critically, the Chinese government recently announced only SA-compatible 5G smartphones will be allowed to have access to the 5G network in China, starting January 1, 2020. This signals that the focus of future 5G network development will be centered around SA mode instead of this year’s NSA mode. Over 90% of China’s 5G network was NSA mode in 2019. Building a 5G SA network will take longer and cost more.  The market expects China to build as much as 1 million units of 5G base stations in 2020. Even if this goal is achieved, it only accounts for about 11% of total Chinese base stations. Chart 6Chinese Smartphone Sales: Still In Contraction Chinese Smartphone Sales: Still In Contraction Chinese Smartphone Sales: Still In Contraction Lack of variety of SA-compatible 5G-phone models. There are also limited options for SA-compatible 5G smartphones models. So far, even though Huawei, Xiaomi, Vivo, Oppo, ZTE and Samsung have all released 5G smartphones, only models from Huawei work under SA networks.2 All others only work under the NSA network. Hence, the variety of SA-compatible 5G phone models is very limited. This will likely delay sales of 5G phones in China. Many more models of SA-compatible 5G smartphones will likely be released only in the second half of next year, which may both drive down 5G smartphone prices and attract more buyers. Consumer spending slowdown: 4G smartphones can meet the needs of the majority of users, and most users have purchased a new phone within the past three years. With elevated economic uncertainty and slowing income growth, a larger proportion of people in China may decide to delay upgrading from 4G-phones to much more expensive 5G ones. This echoes a continuing decline in Chinese smartphone sales (Chart 6). Overall, from Chinese consumers’ perspective, a 5G phone in 2020 will be a nice-to-have, but not a must-have. Given all the aforementioned factors, our best guess for 2020 Chinese 5G smartphone shipments is 40-60 million units, with a larger proportion occurring in the second half of the year.  From Chinese consumers’ perspective, a 5G phone in 2020 will be a nice-to-have, but not a must-have. As China is much more aggressive in moving to 5G network adaptation than other large economies, we share industry experts’ forecasts that China will account for 50% of total global 5G shipments. Provided our estimate for China is about 50 million units, our global forecast for 5G phone shipments in 2020 comes to about 100 million units worldwide. This is substantially lower than industry and analyst average estimates of 180 million units (see Table 1 on page 4). Notably, rising 5G smartphone sales will cannibalize some 4G-phone demand. Consequently, aggregate demand for semiconductors will not grow, but the share of high-valued-added chips in the overall product mix will rise. Bottom Line: The penetration pace of 5G smartphones will be meaningfully slower than both the semiconductor producers and analysts expect. Most likely, a meaningful recovery in global aggregate smartphone sales will not occur over the next six months. We suspect the positive impact of 5G phone sales will be felt by global semiconductor producers largely in the second half of 2020. Semiconductor Demand Beyond 5G In terms of end usage, except smartphones, the top five end uses for semiconductors are personal computers (PCs) (12%), servers (11%), diverse consumer products (12%), automotive (10%), and industrial electronics (9%). Structural PC demand is down, but sales have been more or less flat in the past three years (Chart 7). Next year, commercial demand may accelerate as enterprises work through the remainder of their Windows 10 migration. However, household demand is still facing strong competition from tablets. Overall, we expect PC demand to remain stagnant. Global server shipments sank deeper into contraction in the second quarter of this year due to a slowdown in purchasing from cloud providers and hyperscale customers. They may stay in moderate contraction over the next six months as global economic uncertainty remain elevated, which may discourage enterprises’ investment plans (Chart 8). Chart 7Structural PC Demand Is Stagnant And Will Remain So In 2020 Global PCs Sales: Deeply Saturated Structural PC Demand: Remain Stagnant In 2020 Global PCs Sales: Deeply Saturated Structural PC Demand: Remain Stagnant In 2020 Chart 8Global Server Shipments: A Moderate Contraction In 2020 Global Server Shipment: Are In Contraction Global Server Shipments: Moderate Contraction In 2020 Global Server Shipment: Are In Contraction Global Server Shipments: Moderate Contraction In 2020 Chart 9Automotive-Related Semiconductor Demand: A Moderate Growth Ahead Automotive-Related Semiconductor Demand: A Moderate Growth Ahead Automotive-Related Semiconductor Demand: A Moderate Growth Ahead Chinese auto sales – about 30% of the world total – will likely stage a rate-of-change improvement, moving from deep to mild contraction or stagnation over the next six months.3 Increasing penetration of new energy vehicles and continuing 5G deployment may still result in moderate growth in auto-related semiconductor demand (Chart 9). Semiconductor demand from diverse consumer products slightly declined in the third quarter, with robust growth in tablets, eReaders and portable navigation devices, and contraction in all other subsectors including TV sets, gaming, printers and images, cameras and set-top boxes (Chart 10). This may remain in slight contraction or stagnation over the next three to six months. Automation and 5G deployment will likely continue to increase semiconductor sales in the industrial sector (Chart 11).  Chart 10Semiconductor Demand From Consumer Products: A Slight Contraction Or Stagnation Ahead Semiconductor Stocks: Is The 5G Hype Warranted? Semiconductor Stocks: Is The 5G Hype Warranted? Chart 11Industrial Semiconductor Demand: More Upside Ahead Industrial Semiconductor Demand: More Upside Ahead Industrial Semiconductor Demand: More Upside Ahead   Chart 12Memory Prices Still Signal Sluggish Semiconductor Demand Memory Prices Still Signal Sluggish Semiconductor Demand Memory Prices Still Signal Sluggish Semiconductor Demand Overall, demand recovery has not yet begun. The lack of price recovery in DRAM prices after 18 months of declines and still-low NAND prices are also signaling sluggish semiconductor demand (Chart 12). Bottom Line: Odds are that global semiconductor demand in sectors other than smartphones will show improvement in terms of rate of change, but will still likely be flat in 2020. TSMC Sales: A Harbinger Of Industry Recovery? TSMC, the world’s biggest semiconductor company, posted a revival in sales over four consecutive months from June to September. Do TSMC sales lead global semiconductor sales? The answer is not always. TSMC sales do not always correlate well with global semiconductor sales (Chart 13). For example, TSMC sales diverged from global semiconductor sales in 2017-‘18 and 2013-‘14. So what are the reasons for strong increase in TSMC sales? First, it reflects market share rotation in the global smartphone market in favor of smartphone producers that use TSMC-fabricated chips. Chart 13TSMC Sales Do Not Always Lead Global Semiconductor Sales TSMC Sales Do Not Always Lead Global Semiconductor Sales TSMC Sales Do Not Always Lead Global Semiconductor Sales Demand from the global smartphone sector contributes to almost half of TSMC’s total revenue. Apple and Huawei are TSMC’s two top customers. The most recent report from market research firm Canalys shows that while Apple’s smartphone shipments declined 7% year-on-year last quarter, Huawei’s shipments soared 29%.4 Combined, smartphone shipments from these two companies still jumped nearly 12% year-on-year in the third quarter of the year. This has increased their market share in the global smartphone market to 31% now from 28% a year ago. Second, rising TSMC sales also reflect market share rotation in the global server market, in particular rising shipments and growing market share of servers using AMD high-performing-computing (HPC) chips instead of Intel ones. AMD’s 7nm Epyc CPU, launched this August and manufactured by TSMC, has been taking share from Intel in the global server market. This has driven the increase in TSMC’s revenue from the HPC sector.  Third, the share of value-added products (high-end chips) in TSMC’s product mix has been rising rapidly. TSMC’s share of revenue from 7nm technology jumped from 21% to 27% in the third quarter, as most of Apple’s and Huawei’s chips and all of AMD’s Epyc CPUs are 7nm-based. Back in the third quarter of 2018, TSMC’s 7nm business only accounted for 11% of its total revenue. Chart 14Both TSMC Sales And Taiwanese PMI Could Continue To Improve While Global Semiconductor Sales Remain In Contraction Both TSMC Sales And Taiwanese PMI Could Continue To Improve While Global Semiconductor Sales Remain In Contraction Both TSMC Sales And Taiwanese PMI Could Continue To Improve While Global Semiconductor Sales Remain In Contraction Finally, although internet of things (IoT) and automotive chips only account for 9% and 4% of TSMC’s total share of revenue respectively, strong growth in both segments –33% year-on-year in IoT and 20% year-on-year in automotive – indeed shows exceptional demand in these two sectors in a weakening global economic environment. As IoT and automotive development will highly rely on global 5G infrastructure development, their impact will be meaningful once the global 5G network becomes well advanced and widely installed. To conclude, while a 40% boost in TSMC’s capital spending indeed paints a positive picture on global semiconductor demand over the longer term, rising TSMC sales do not mean an imminent and strong recovery in the global semiconductor sector is in the works. Huawei is the global 5G technology leader and the major supplier in both 5G-network equipment and 5G smartphones; the company will be a major revenue contributor to TSMC. As Huawei will likely place more orders to TSMC for chip fabrication, this will likely result in further improvement in TSMC’s sales (Chart 14). Bottom Line: Rising TSMC sales do not necessarily herald an imminent and robust cyclical recovery in the global semiconductor sector.  Investment Conclusions Global semiconductor stock prices have been front running a recovery that has not yet begun. In addition, there is still uncertainty about the technology aspect of US-China trade negotiations. The US will likely continue to have Huawei and other Chinese high-tech companies on its trade-ban list – its so-called Entity List. TSMC sales do not always correlate well with global semiconductor sales. Notably, global semiconductor sales and profits are still in deep contraction, while share prices are at all-time highs (Chart 15). As a result, semiconductor stocks’ multiples have spiked to their previous highs (Chart 16). Chart 15Semiconductor Companies Profits: Still In Deep Contraction Semiconductor Companies Profits: Still In Deep Contraction Semiconductor Companies Profits: Still In Deep Contraction Chart 16Elevated Semiconductor Stocks Multiples Elevated Semiconductor Stocks Multiples Elevated Semiconductor Stocks Multiples     While it is common for share prices to rally ahead of a business cycle/profit revival, we believe a true recovery will only emerge in spring 2020, and it will initially be much more subdued than industry watchers and investors expect. In the near term, strong momentum could still push semiconductor stock prices higher. However, the reality will then set in and there will be an air pocket before a more sustainable bull market emerges.   Our US Equity Investment Strategy earlier this week downgraded S&P semiconductor equipment companies to underweight and put the S&P Semiconductors Index on a downgrade alert.5 Their newly created top-down semiconductor profit growth model warns that an earnings recovery is not yet imminent (Chart 17). For EM-dedicated equity managers, we have been neutral on Asian semiconductor sectors. We continue to recommend a market-weight allocation to Taiwan’s overall market, while we are upgrading the Korean technology sector from a neutral allocation to overweight. Korean semiconductor stocks have rallied much less than their global peers. Hence, the risk of a major relapse is lower. Given that we have been overweight non-tech Korean stocks, upgrading tech stocks to overweight means we will be overweight the KOSPI within the EM equity benchmark (Chart 18). Chart 17Semiconductor Earnings Recovery: Not Imminent Semiconductor Earnings Recovery: Not Imminent Semiconductor Earnings Recovery: Not Imminent Chart 18Upgrade Korean Tech Stocks And Overweight KOSPI Within EM Upgrade Korean Tech Stocks And Overweight KOSPI Within EM Upgrade Korean Tech Stocks And Overweight KOSPI Within EM   Meanwhile, we remain long the Bloomberg Asia-Pacific Semiconductor Index and short the S&P 500 Semiconductor Index. This trade has produced a 7% gain since its initiation on June 13, 2019. The Bloomberg Asia-Pacific Semiconductor index has 12 stocks. Samsung and TSMC account for 38% and 37% of the index, respectively. The S&P 500 Semiconductor Index has 13 stocks. Intel, Broadcom, Texas Instruments and Qualcomm are the top five constituents, together accounting for nearly 77% of the index. Although the US and China may reach a temporary trade deal, the US will continue to restrict sales of tech products and high-end semiconductors to China. As a result, these US semiconductor companies, most of which are IC designing companies, will likely experience a more subdued than expected recovery in sales. Ellen JingYuan He Associate Vice President ellenj@bcaresearch.com Footnotes   1 Please see Emerging Markets Strategy Special Report "The Global Semiconductor Sector: Is A Cyclical Upturn Imminent?" dated June 13, 2019, available at ems.bcaresearch.com 2 https://www.guancha.cn/ChanJing/2019_09_21_518748.shtml http://www.cac.gov.cn/2019-10/23/c_1573361796389322.htm 3 Please see Emerging Markets Strategy Special Report "Chinese Auto Demand: Time For A  Recovery?" dated October 17, 2019, available at ems.bcaresearch.com 4 https://www.canalys.com/analysis/smartphone+analysis 5 Please see US Equity Strategy Special Report "Defying Gravity," dated November 4, 2019, available at uses.bcaresearch.com

Highlights The odds of a cyclical upturn in the global semiconductor sector over the next three to six months are low. Global semiconductor demand will continue to decline due to contracting demand for smartphones, automobiles, personal computers (PCs), and servers. Global semiconductor stocks are still facing considerable downside in absolute terms. We recommend going long Asian semiconductor stocks versus the U.S. S&P 500 semiconductor index. Dedicated EM equity portfolios should stay neutral on the Taiwanese bourse and Korean technology sector relative to the overall EM benchmark. Feature Chart 1 shows share prices of the global semiconductor sector and global semiconductor sales in the past two decades. Chart 1Global Semiconductor Market: Sales & Share Prices Global Semiconductor Market: Sales & Share Prices Global Semiconductor Market: Sales & Share Prices Was last December’s trough in global semiconductor equity prices the ultimate bottom in this cycle? The odds are in favor of a continued contraction in global semiconductor sales and further downside in semiconductor share prices over the next three to six months. Cycle-On-Cycle Analysis Semiconductor sales experienced five recessions over the past 20 years. Table 1 illustrates the peak-to-bottom percentage decline in nominal global semiconductor sales and the magnitude of the drop in global semiconductor share prices in U.S. dollar terms during these five cyclical downturns in this industry. It also indicates the duration of each downturn and the number of months that semiconductor stocks led the bottom in global semiconductor sales. Table 1Key Statistics Of Five Cyclical Downturns In Global Semiconductor Market The Global Semiconductor Sector: Is A Cyclical Upturn Imminent? The Global Semiconductor Sector: Is A Cyclical Upturn Imminent? The current shrinkage of semiconductor sales is worse than the 2011-12 and 2015 downturns. Yet, it is still smaller than the magnitude during the 2008 Great Financial Crisis and the 2001 tech bubble bust. The revenue of semiconductor companies has so far contracted by 24%, which is disproportionally more than the decline in share prices of these companies. The global semiconductor equity index is only 14% below its March 2018 high. It appears as though the market is expecting a quick recovery in semiconductor sales. As per Table 1, in the downturns of 2008, 2011 and 2015, global semiconductor stocks all bottomed before the bottom of global semiconductor sales. Only in the 2001 episode, stock prices bottomed eight months after the bottom in sales. In the current cyclical downturn, global semiconductor sales have so far had only four months of growth contraction,1 far less than the 13-16 months experienced in all the past four cycles. All in all, we would lean against the market’s expectation of an imminent recovery in the semiconductor cycle. The demand downturn will last another three to six months and share prices are facing major headwinds. Global Semiconductor Demand Semiconductor sales are in contraction across countries and regions (Chart 2). In April – before President Trump’s tweet on imposing new import tariff on China, global semiconductor sales growth sank to a negative 15% year-on-year. The short-term (three-to-six month) outlook for global semiconductor demand remains dismal. Chart 3 shows global semiconductor revenue breakdown in terms of end usage. Mobile phones account for the largest share (29%) of the market, followed by PCs (12%), miscellaneous consumer products (12%), and servers (11%). All of these major demand sources are under downward pressure: Smartphone Sales Global smartphone sales are shrinking (Chart 4). According to the International Data Corporation (IDC), global smartphone shipments declined 6.6% year-on-year in volume terms in the first quarter of this year, worse than last year’s 4.4% drop. Chart 2Semiconductor Sales Are In Contraction Across Countries Semiconductor Sales Are In Contraction Across Countries Semiconductor Sales Are In Contraction Across Countries  In the current cyclical downturn, global semiconductor sales have so far had only four months of growth contraction, far less than the 13-16 months experienced in all the past four cycles. Chart 3 Chart 4Global Smartphone Sales: Contracting Global Smartphone Sales: Contracting Global Smartphone Sales: Contracting We expect smartphone shipments to continue contracting in the second half of this year. Major markets such as mainland China and advanced economies have entered the saturation phase of mobile-phone demand. For example, U.S. shipments were down 15% year-on-year in the first quarter due to near-full market penetration. In China, smartphone sales have shown signs of stabilization (Chart 5). However, this is probably temporary and has been driven by the boom in Huawei smartphone sales in China since early this year. The incredible 50% year-on-year growth of Huawei smartphone sales in the first quarter is not sustainable. While global sales of Huawei smartphones increased by 20 million units, total global smartphone sales of all brands fell by 22 million units (Chart 6). The U.S. punitive actions towards China and Huawei have also instigated nationalism in China. This has triggered a Chinese buying-spree of the Huawei smartphone. Chart 5Chinese Smartphone Sales: Temporary Stabilization Chinese Smartphone Sales: Temporary Stabilization Chinese Smartphone Sales: Temporary Stabilization Chart 6 Yet, this has probably reduced the number of potential Chinese smartphones buyers in the near future. After all, many buyers likely made the purchase earlier than otherwise planned in the absence of a trade war. Although Samsung, Huawei, OnePlus, Xiaomi, Motorola, LG, and ZTE have either released or will release their 5G phones this year, the sales growth from 5G phones will not be able to offset the loss in 2G, 3G and 4G phone sales, at least not in 2019. The IDC estimated that 5G phones would only account for about 0.5% of the market share this year. 5G will likely only begin affecting overall semiconductor demand next year, when they account for a larger share of smartphone sales.   Huawei is the market leader in 5G technology. The U.S. boycott of Huawei will likely continue. This will only slow the pace of 5G phone adoption and the development of 5G networks worldwide. On balance, global smartphone demand may only recover next year. Server Demand Global server shipments also experienced a 5% contraction in volume terms in the first quarter of this year, according to IDC (Chart 7). The outlook for the rest of 2019 does not look promising. Global server demand will likely remain in contraction in the second half of this year. Many hyperscale data centers have already purchased considerable amounts of severs in advance of the trade war to avoid tariffs.2 Meanwhile, the escalation in the U.S.-China confrontation has increased economic uncertainties. This may delay potential datacenter investments. Decelerating 5G network development worldwide due to the U.S. ban on Huawei will also tend to discourage new datacenter and cloud services projects. This is because the 5G technology enables datacenter and cloud services to experience a huge improvement in terms of data transfer speeds, latency, connectivity, capacity, reliability and mobility. Chart 7Global Server Shipment: Are In Contraction Global Server Shipment: Are In Contraction Global Server Shipment: Are In Contraction Personal Computers (PC) PCs sales are also in contraction (Chart 8). PC demand has entered into the deep-maturation phase while facing strong competition from tablets and laptops. Auto Sales Global auto sales also sank by 5% in April from a year ago, registering the worst contraction since 2009 (Chart 9). Chart 8Global PCs Sales: Deeply Saturated Global PCs Sales: Deeply Saturated Global PCs Sales: Deeply Saturated Chart 9Global Auto Sales Are In Contraction As Well Global Auto Sales Are In Contraction As Well Global Auto Sales Are In Contraction As Well Regarding auto demand, the Chinese government may continue to implement more supportive policies to stimulate car sales in China. However, we believe the recovery will be delayed. The government has already implemented a number of policies to lift domestic car sales since late January, including providing subsidies to encourage new energy vehicle sales, to promote auto sales in rural areas, and to increase auto replacement. The central government recently loosened auto sales restrictions in the first tier cities of Guangzhou and Shenzhen that have restrictive auto sales policies. However, all of these policies have failed to lift Chinese domestic car sales out of deep contraction. The key reason has been a diminishing willingness to spend among Chinese consumers, as suggested by falling households’ marginal propensity to consume (Chart 10). Bottom Line: Global semiconductor demand growth will likely remain weak and will fail to recover in the second half of this year. The basis is that its major upstream markets (smartphone, servers, PCs and automobiles) are all facing cyclically declining demand. Chart 10Chinese Consumers: Diminishing Willingness To Consume Chinese Consumers: Diminishing Willingness To Consume Chinese Consumers: Diminishing Willingness To Consume Inventories And Prices Chart 11 shows the semiconductor supply chain illustrating the process of manufacturing semiconductors starting with silicon wafers and up to final electronic products. Chart 11 Box 1 explains the role of key segments and players along the supply chain. Box 1 A Brief Explanation Of The Key Segments/Players Of The Supply Chain Both integrated device manufacturers (IDM) and foundries are at the center of the supply chain, responsible for chip manufacturing. In terms of semiconductor sales revenue, Samsung, Intel and SK Hynix are the world’s top three IDM companies and TSMC, Global Foundries and United Microelectronics Corp (UMC) are the world’s top three foundries. While IDMs cover most of the process from IC design, chip fabrication, assembly, testing and packaging, IDM companies still have to purchase raw materials and equipment for the chip-making process. Foundry companies receive orders from IC designing companies like Qualcomm, Nvidia, and Huawei HiSilicon, then purchase needed raw materials and equipment to proceed in the chip-manufacturing process. Both IDMs and foundries can either outsource the tasks of semiconductor assembly and testing or perform them on their own. The final semiconductor products will be used in electronics products, such as smartphones, computers, home appliances, automobiles, etc. Global semiconductor demand growth will likely remain weak and will fail to recover in the second half of this year. The basis is that its major upstream markets (smartphone, servers, PCs and automobiles) are all facing cyclically declining demand. In a typical business cycle, a cyclical downturn begins with a slump in demand for final electronic products (upstream demand). This leads to falling semiconductor sales. As a result, inventory buildup will occur across most of the segments along the semiconductor supply chain. Chipmakers: Producers’ semiconductor inventory in both Taiwan and Korea have reached either a record high or a near-record high (Chart 12). The installed wafer capacities at these two countries are the world’s largest, together accounting for 43% of total global wafer capacity. In addition, the inventory of some major electronic parts and components have also increased considerably in Taiwan (Chart 13). This also implies weaker demand for semiconductor raw materials. Chart 12Chipmakers: A Rapid Buildup In Inventory Chipmakers: A Rapid Buildup In Inventory Chipmakers: A Rapid Buildup In Inventory Chart 13Rising Inventory Of Some Major Electronic Parts And Components Rising Inventory Of Some Major Electronic Parts And Components Rising Inventory Of Some Major Electronic Parts And Components Raw material suppliers: Silicon wafer is the indispensable raw material required in the chip manufacturing process. Japanese companies account for over half of global silicon wafer supply.Chart 14 shows that silicon wafer inventory in Japan has had a significant buildup in volume terms since late last year. Importantly, it is not declining yet. Chart 14Silicon Wafer Inventory: A Significant Buildup As Well Silicon Wafer Inventory: A Significant Buildup As Well Silicon Wafer Inventory: A Significant Buildup As Well Outsourced semiconductor assembly and test (OSAT) providers: Both Singapore and Thailand are OSAT providers while they also manufacture, assemble and export electronic products. Both countries are closer to the downstream side of the semiconductor supply chain. Semiconductor inventory at these two countries has also jumped to a record high (Chart 15).   Chart 15Singapore and Thailand: Record High Semiconductor inventory Singapore and Thailand: Record High Semiconductor inventory Singapore and Thailand: Record High Semiconductor inventory   Importantly, a marginal improvement in demand will tend to support spot prices. For example, in the memory chip market, falling prices denote weak demand relative to excess supply. When prices of DRAM and NAND start to form a bottom or decisively move up, this may indicate the arrival of a cyclical upturn. So far, both DRAM and NAND prices are continuing to fall (Chart 16). In addition, the prices of silicon wafer – the most important raw material used in the chip-making process – have declined in the first half of this year.3  Chart 16Still Falling Memory Chip Prices Still Falling Memory Chip Prices Still Falling Memory Chip Prices Chart 17Deflating DRAM Prices Suggest Downside Risks To Korean Tech Stocks Deflating DRAM Prices Suggest Downside Risks To Korean Tech Stocks Deflating DRAM Prices Suggest Downside Risks To Korean Tech Stocks In short, prices are the most important variable to monitor. Chart 17 exhibits the high correlation between DRAM prices and the Korean technology sector stock prices. Falling DRAM prices suggest downside risks to technology stocks in Korea. Samsung accounts for about 65% of Korea’s tech index and 27% of the overall Korean equity index. Memory chips accounted for 68% of Samsung’s operating profits in the first quarter of this year.   Bottom Line: There has been involuntary inventory accumulation along the entire supply chain of semiconductors. This and ongoing price deflation among various types of semiconductors foreshadow a downbeat near-term outlook. The Interpretation Of Some Positive Developments There have been some positive developments in the past several months. Taiwanese PMI new orders diffusion index in the electronics sector jumped out of deep contraction to reach zero, and Chinese semiconductor imports halted their decline in both volume and value terms (Chart 18). The improvement in the aforementioned data was probably mainly due to large semiconductor purchases by China to hedge the rising risk of U.S. blocking China’s technological development (Chart 19). Chart 18Some Positive Development Some Positive Development Some Positive Development Chart 19China: More Semiconductors Purchases Before The Tariff And U.S. Huawei Ban? China: More Semiconductors Purchases Before The Tariff And U.S. Huawei Ban? China: More Semiconductors Purchases Before The Tariff And U.S. Huawei Ban? Besides, Huawei smartphone sales have been booming, which we deliberated on page 5, could have been responsible for the improvement in these data. This one-off surge will likely dwindle going forward. Investment Conclusions We remain negative on Asian semiconductor share prices in absolute terms. A continued contraction in global semiconductor sales will further squeeze their profits. In relative terms, we are neutral on the Asian semiconductor sector: we continue recommending market-weight allocation to Taiwan’s overall market and the Korean technology sector within the EM equity benchmark. As a new trade, we recommend going long Asian semiconductor stocks and short the S&P 500 semiconductor index over the next three to six months (Chart 20). The Bloomberg Asia Pacific semiconductor index has nine stocks. Samsung and TSMC account for 42% and 38% of the index, respectively. There has been involuntary inventory accumulation along the entire supply chain of semiconductors. This and ongoing price deflation among various types of semiconductors foreshadow the downbeat near-term outlook. Samsung will likely benefit from the U.S. ban on Huawei in the smartphone sector outside of China. In addition, Samsung will win some market share from Apple as the latter does not have a 5G phone to release this year. These positive factors may partially offset the negative impact from falling memory prices and demand on Samsung. The S&P 500 semiconductor index has 13 stocks. Intel, Broadcom, Texas Instruments and Qualcomm are the top five constituents, together accounting for nearly 70% of the index. Most of these companies are IC designing companies, which will likely suffer as Chinese demand for their products shrink due to the U.S. administration's ban on Huawei. This position will also benefit from U.S. dollar appreciation. A firm dollar will hurt profits of U.S. semiconductor stocks. In turn, currency depreciation in Korea and Taiwan will on the margin benefit Asian semiconductor stocks. Chart 20Recommend Long Asia Pacific Semiconductor Stock Vs. S&P 500 Semiconductor Index Recommend Long Asia Pacific Semiconductor Stock Vs. S&P 500 Semiconductor Index Recommend Long Asia Pacific Semiconductor Stock Vs. S&P 500 Semiconductor Index Chart 21The 2015 Experience The 2015 Experience The 2015 Experience Chart 21 shows that global foundry companies outperformed global IC designing companies during the final phase of the 2015 cyclical downturn. Odds are that these dynamics will play out in this downturn as well. Finally, the relative performance of Asian semiconductor stocks versus U.S. ones is oversold and might stage some sort of mean reversion (Chart 20). Ellen JingYuan He, Associate Vice President ellenj@bcaresearch.com Footnotes 1      Please note that here the calculation for “the number of months of the growth contraction” is different from the one for the “peak-to-bottom duration” in Table 1. “The number of months of the growth contraction” equals the number of months when semiconductor sales year-on-year growth is negative. 2      https://marketrealist.com/2019/05/nvidias-data-center-revenue-inference-rendering-and-edge/ 3      http://www.sohu.com/a/300386061_132567, http://news.moore.ren/industry/104958.htm   Equity Recommendations Fixed-Income, Credit And Currency Recommendations

Highlights So what? Quantifying geopolitical risk just got easier. Why?   In this report we introduce 10 proprietary, market-based indicators of country-level political and geopolitical risk. Featured countries include France, U.K., Germany, Italy, Spain, Russia, South Korea, Taiwan, Turkey, and Brazil. Other countries, and refinements to these beta-version indicators, will come in due time. We remain committed to qualitative, constraint-based analysis. Our GeoRisk Indicators will help us determine how the market is pricing key risks, so we can decide whether they are understated or overstated. Feature For the past three months we have been tracking a “Witches’ Brew” of political risks that threaten the late-cycle bull market. Some of these risks have abated for the time being: the Fed is on pause, China’s stimulus has surprised to the upside, and Brexit has been delayed. Other risks we have flagged, however, are heating up: Iran And Oil Market Volatility: Surprisingly the Trump administration has chosen not to extend oil sanction waivers on Iran from May 2, putting 1.3 million barrels per day of oil on schedule to be removed from international markets by an unspecified time.  It remains to be seen how rapidly and resolutely the administration will enforce the sanctions on specific allies and partners (Japan, India, Turkey) as well as rivals (China, others). Because the decision coincides with rising production risks from renewed fighting in Libya and regime failure in Venezuela, we expect President Trump to phase in the new enforcement over a period of months, particularly on China and India. But official rhetoric is draconian. Hence the potential for full and immediate enforcement is greater than we thought. In the short term, individual political leaders, and very powerful nations like the United States, can ignore material economic and political constraints. Since the Trump administration’s decision exemplifies this point, geopolitical tail risks will get fatter this year and next. Global oil price volatility and equity market volatility will increase with sanction enforcement actions and retaliation. We would think that Trump’s odds of reelection will marginally suffer, though for now still above 50%, as any full-fledged confrontation with Iran will raise the chances of an oil price-induced recession. U.S.-EU Trade War: Neither the Trump administration nor the U.S. has a compelling interest in imposing Section 232 tariffs on imports of autos and auto parts. Nevertheless the risk of some tariffs remains high – we put it at 35% – because President Trump is legally unconstrained. The decision is technically due by May 18 but Economic Council Director Larry Kudlow has said Trump may adjust the deadline and decide later. Later would make sense given the economic and financial risks of the administration’s decision to ramp up the pressure on Iran.1 But the risk that tariffs will pile onto a weak German and European economy will hang over investors’ heads. U.S.-China Talks Not A Game Changer: The ostensible demand that China cease Iranian oil imports immediately and the stalling of U.S. diplomacy with North Korea are not conducive to concluding a trade deal in May. We have highlighted many times that strategic tensions will persist even if Beijing and Washington quarantine these issues to agree to a short-term trade truce. The June 28-29 G20 meeting in Japan remains the likeliest date for a summit between Presidents Trump and Xi Jinping, but even this timeframe could be too optimistic. Continued uncertainty or a weak deal will fail to satisfy financial markets expecting a very positive outcome.   With a 70% chance that U.S. tariffs on China will not increase this year and, contingent on a U.S.-China deal, only a 35% chance that the U.S. slaps tariffs on German cars, we sound optimistic to some clients. But the Trump administration’s decision on Iran is highly market-relevant and portends greater volatility. We expect to see a geopolitical risk premium creep higher into oil markets as well as a greater risk of “Black Swan” events in strategically critical or oil-producing parts of the Middle East. There is limited research devoted to quantifying geopolitical risk. We are late in the business cycle and President Trump has emphatically decided to increase rather than decrease geopolitical risk. Quantifying Geopolitical Risk Geopolitical analysis has taken a bigger role in investors’ decision-making over the last decade. Surveys show that geopolitical risks rank among global investors’ top concerns overall. In the oft-cited Bank of America Merrill Lynch survey, geopolitical and related issues have dominated the “top tail risk” responses for the past half-decade (Chart 1). In other surveys, the most worrisome short-term risks are mostly political or geopolitical in nature, ranking above socio-economic and environmental risks (Chart 2). Chart 1 Chart 2 Despite this high level of concern, there is limited research devoted to quantifying geopolitical risk. Isolating and measuring the range of risks under this umbrella term remains a challenge. As such, for many investors, geopolitics remains an ad hoc, exogenous factor that is often mentioned but rarely incorporated into portfolio construction. For the past four decades the predominant ways of measuring political or geopolitical risk have been qualitative or semi-qualitative. The Delphi technique, developed on the basis of low-quality data sets in social sciences, relies on pooled expert opinions.2 Independently selected experts are asked to provide risk assessments and their responses are then interpreted by analysts to create a measure of risk. Another semi-qualitative method of measuring geopolitical risk ranks countries according to a set of political and socio-economic variables. These variables – such as governance, political and social stability, corruption, law and order, or formal and informal policies – are extremely important but inherently difficult to quantify.3 These results are useful but suffer from dependency on expert opinion, data quality, and institutional biases. More importantly, these methods are slow to react to breaking events in a rapidly changing world. The same goes for bottom-up assessments using political intelligence. The weakness of these methods is that it is highly unlikely that they will produce statistically significant estimates of risk. The odds of getting a “silver bullet” insight from a “key insider” are decent for simple political systems, but not in the complex jurisdictions that host the vast majority of global, liquid investments. Quantitative approaches to measuring geopolitical risk have since become more widespread. The most prominent method is based on quantifying the occurrence of words related to political and geopolitical tensions that appear in international newspapers. These word-counts typically include terms like “terrorism,” “crisis,” “war,” “military action,” etc. As a result, the indices reflect incidents of physical violence or other “Black Swan” events that may not have direct relevance to financial markets. Moreover, while news-based indices accurately capture dramatic one-time peaks at the time of a crisis, they are largely flat aside from these, as they rely on popular topics rather than underlying structural trends (Chart 3). They fail to capture geopolitical developments associated with electoral cycles, protest movements, paradigm shifts in economic policy, or other policy changes.4 Notice, for instance, that the fall of the Soviet Union in late 1991 and the resulting chaos in Russia and many other parts of the emerging world hardly register in Chart 3. Chart 3News-Based Indices Only Capture Crisis Peaks, Not Geopolitical Developments News-Based Indices Only Capture Crisis Peaks, Not Geopolitical Developments News-Based Indices Only Capture Crisis Peaks, Not Geopolitical Developments Introducing BCA’s GeoRisk Indicators The past 70 years have taught BCA Research to listen and respect the market. Why would we suddenly follow the media instead? Most quantitative geopolitical indicators begin with the premise that journalists and the news-reading public have accurately emphasized the most relevant risks and uncertainties. They proceed to quantify the terms of these assessments with increasingly sophisticated methods. This approach solves only part of the puzzle. News-based indices ... fail to capture geopolitical developments associated with underlying policy changes. At BCA Geopolitical Strategy, we aim to generate geopolitical alpha.5 This means identifying where financial media and markets overstate or understate geopolitical risks. We do not primarily aim to predict events or crises. As such, traditional news-based indicators that capture only major events, even those ex post facto, are of little relevance to our analysis. What is needed is a better way to quantify how the market is calculating risks. We start with a simple premise: the market is the greatest machine ever created for gauging the wisdom of the crowd. Furthermore, it puts its money where its predictions are, unlike other methods of geopolitical risk quantification which have no “value at risk.” Chart 4USD/RUB Captures Geopolitical Risk In Russia... USD/RUB Captures Geopolitical Risk In Russia... USD/RUB Captures Geopolitical Risk In Russia... To this end, we have introduced market-based indicators over the years that rely on currency movements, which are often the simplest and most immediate means of capturing the process of pricing risk. In 2015, for instance, we introduced an indicator that measures Russia’s geopolitical risk premium (Chart 4). It is constructed using the de-trended residual from a regression of USD/RUB against USD/NOK and Russian CPI relative to U.S. CPI. We can show empirically that it captures geopolitical risk priced into the ruble, as the indicator increases following critical incidents. These include the downing of Malaysian Airlines Flight 17 over eastern Ukraine in 2014; the warnings that Russia aimed to stage a “spring offensive” in Ukraine in 2015; Russian military intervention in the Syrian Civil War later that year; and the poisoning of former intelligence agent Sergei Skripal in the U.K. in 2018 and subsequent tensions. Using similar methods, we created a proxy to capture geopolitical risk in Taiwan, based on USD/JPY and USD/KRW exchange rates and relative Taiwanese/American inflation (Chart 5). The indicator tracks well with previous cross-strait crises. It jumped upon Taiwan’s election of President Tsai Ing-wen and her pro-independence government in January 2016 – and this was well before any tensions actually flared. It even registered a small increase upon her controversial phone call congratulating Donald Trump upon winning the U.S. election. Chart 5...And USD/TWD Captures Geopolitical Risk In Taiwan ...And USD/TWD Captures Geopolitical Risk In Taiwan ...And USD/TWD Captures Geopolitical Risk In Taiwan This year we have expanded on this work, constructing a set of ten standardized GeoRisk Indicators for five developed economies and five emerging economies: U.K., France, Germany, Spain, Italy, Russia, Turkey, Brazil, Korea, and Taiwan. Indicators for the U.S., China, and others will be rolled out in a future report. These indicators attempt to capture risk premiums priced into the various currencies – except for Euro Area countries, where the risk is embedded in equity prices. In each case, we look at whether the relevant assets are decreasing in value at a faster rate than implied by key explanatory variables. The explanatory variables consist of (1) an asset that moves together with the dependent variable while not responding to domestic geopolitical risks, and (2) a variable to capture the state of the economy. This set of indicators differs from our earlier indicators in the following ways: We aim to create a simple methodology that we can apply consistently to all countries, both in the DM and EM universes. We therefore omitted using regression models that can prove to be quite whimsical. Instead, we simply looked at the deviation of the dependent variable from the explanatory variables, all in expanding standardized terms, to create the GeoRisk proxy. We wanted an indicator that would immediately respond to priced-in risks, so we opted for a daily frequency rather than the weekly frequency we used in our initial work. To get as accurate of a signal as possible, we use point-in-time data. Since economic data tends to be released with a one-to-two-month lag, we lagged the economic independent variable to correspond to its release date. All ten indicators are shown in the Appendix. Across all countries, they track well with both short-term events and long-term trends in geopolitical risk. In the case of France, for example, the indicator steadily climbs during the period of domestic tensions and protests in the early 2000s; as the European debt crisis flares up; again during the rise of the anti-establishment Front National and the Russian military intervention in Ukraine; and finally during the U.S. trade tariffs and Yellow Vest protests (Chart 6). Our GeoRisk indicators isolate risks that either originate internally or otherwise affect the country more so than others. Similarly, in Germany, there is a general increase in perceived risk as Chancellor Gerhard Schröder implements structural reforms in the early 2000s; another increase leading up to the leadership change as Angela Merkel is elected Chancellor; another during the global and European financial crises; another during the Ukraine invasion and refugee influx; and finally another with the U.S.-China trade war (Chart 7). Chart 6Our French Indicator Picks Up Domestic And European Unrest Our French Indicator Picks Up Domestic And European Unrest Our French Indicator Picks Up Domestic And European Unrest Chart 7Greater German Risk Amid The Trade War Greater German Risk Amid The Trade War Greater German Risk Amid The Trade War   We have annotated each country’s GeoRisk indicator heavily in the appendix so that readers can see for themselves the correspondence with political events. The indicators are affected by international developments – like the Great Recession – but we have done our best to isolate risks that either originate internally or otherwise affect the country more than other countries. (As a consequence, the Great Recession is muted in some cases.) What are the indicators telling us now? Most obviously, they highlight the extreme risk we have witnessed in the U.K. over the now-delayed March 29 Brexit deadline. We would bet against this risk as the political reality has demonstrated that a “hard Brexit” is very low probability: the U.K. has the ability to back off unilaterally while the EU is willing to extend for the sake of regional stability. In this sense the pound is a tactical buy, which our foreign exchange strategist Chester Ntonifor has highlighted.6 Our U.K. risk indicator has been fairly well correlated with the GBP/USD since the global financial crisis and it suggests that the pound has more room to rally (Chart 8). Chart 8Betting Against A Hard Brexit, the GBP Is A Tactical Buy Betting Against A Hard Brexit, the GBP Is A Tactical Buy Betting Against A Hard Brexit, the GBP Is A Tactical Buy Meanwhile, Spanish risks are overstated while Italy’s are understated. As for the emerging world, Turkish risks should be expected to spike yet again, as divisions emerge within the ruling coalition in the wake of critical losses in local elections and a failure to reassure investors over monetary policy and the currency. Brazilian risks will probably not match the crisis points of the impeachment and the 2018 election, at least not until controversial pension reforms reach a period of peak uncertainty over legislative passage. Both our new Russian indicator and its prototype are collapsing (see Chart 4 above). This captures the fact that we stand at a critical juncture in Russian affairs, where President Putin is attempting to shift focus to domestic stability even as the U.S. and the West maintain pressure on the economy to deter Russia from its aggressive foreign policy. Given that both Putin’s and the government’s approval ratings are low amid rising oil prices, the stage is set for Russia to take a provocative foreign policy action meant to distract the populace from its poor living conditions. Venezuela is the obvious candidate, but there are others. Moscow will want to test Ukraine’s newly elected, inexperienced president; it may also make a show of support for Iran. With Russia equities having rallied on a relative basis over the past year and a half, and with the Iranian waiver decision already boosting oil prices as we go to press, the window of opportunity to buy Russian stocks is starting to close. (We remain overweight relative to EM on a tactical horizon; our Emerging Markets Strategy is also overweight.) Going forward, we will update these risk indicators regularly as needed and publish the full appendix at the end of every month along with our long-running Geopolitical Calendar. We will also fine-tune the indicators as new information comes to light. In other words, here we present only the beta version. We hope that these indicators will help inform investors as to the direction, and even magnitude, of political risks as the market prices them. Our GeoRisk indicators are not predictive, as establishing a trend is not a prediction. The main purpose of this exercise is to answer the critical question, “What is already priced in?” How is the market currently calculating geopolitical risk for a country? After that, it is the geopolitical strategist’s job to unpack this question through qualitative, constraint-based analysis. It is when our qualitative assessments disagree with what is priced in that we can generate geopolitical alpha.    Ekaterina Shtrevensky, Research Analyst ekaterinas@bcaresearch.com Matt Gertken, Vice President Geopolitical Strategist mattg@bcaresearch.com Marko Papic Consulting Editor marko@bcaresearch.com Footnotes 1      See Sean Higgins, “Auto tariffs decision could be delayed, Kudlow says,” Washington Examiner, April 3, 2019, www.washingtonexaminer.com. 2      Norman C. Dalkey and Olaf Helmer-Hirschberg, “An Experimental Application of the Delphi Method to the Use of Experts,” Management Science, Vol. 9, Issue: 3 (April 1963) pp. 458- 467. 3      Darryl S. L. Jarvis, “Conceptualizing, Analyzing and Measuring Political Risk: The Evolution of Theory and Method,” Lee Kuan Yew School of Public Policy Research Paper No. LKYSPP08-004 (July 2008).  William D. Coplin and Michael K. O'Leary, "Political Forecast For International Business," Planning Review, Vol. 11 Issue: 3 (1983) pp.14-23. The PRS Group, “Political Risk Services”™ (PRS) or the “Coplin-O’Leary Country Risk Rating System”™ Methodology. Daniel Kaufmann, Aart Kraay, and Massimo Mastruzzi, “The Worldwide Governance Indicators: Methodology and Analytical Issues,” World Bank Policy Research Working Paper No. 5430 (September 2010). 4      Scott R. Baker, Nicholas Bloom, and Steven J. Davis, “Measuring Economic Policy Uncertainty,” The Quarterly Journal of Economics, Volume 131, Issue 4, November 2016 (July 2016) pp.1593–1636. Dario Caldara and Matteo Iacoviello, “Measuring Geopolitical Risk,” Board of Governors of the Federal Reserve Board, Working Paper (January 2018). 5      Please see BCA Research Geopolitical Strategy Special Report, “Five Myths On Geopolitical Forecasting,” dated July 9, 2018, available at gps.bcaresearch.com. 6      Please see BCA Foreign Exchange Strategy Weekly Report, “Not Out Of The Woods Yet,” April 5, 2019, available at www.bcaresearch.com.   Appendix Appendix France France: GeoRisk Indicator France: GeoRisk Indicator Appendix U.K. U.K.: GeoRisk Indicator U.K.: GeoRisk Indicator Appendix Germany Germany: GeoRisk Indicator Germany: GeoRisk Indicator Appendix Italy Italy: GeoRisk Indicator Italy: GeoRisk Indicator Appendix Spain Spain: GeoRisk Indicator Spain: GeoRisk Indicator Appendix Russia Russia: GeoRisk Indicator Russia: GeoRisk Indicator Appendix Korea Korea: GeoRisk Indicator Korea: GeoRisk Indicator Appendix Taiwan Taiwan: GeoRisk Indicator Taiwan: GeoRisk Indicator Appendix Turkey Turkey: GeoRisk Indicator Turkey: GeoRisk Indicator Appendix Brazil Brazil: GeoRisk Indicator Brazil: GeoRisk Indicator What’s On The Geopolitical Radar? Chart 19      Geopolitical Calendar

Highlights Taiwan’s semiconductor sector is facing both cyclical and structural headwinds. Semiconductor exports will continue to contract over the next six months or so, on retrenching global demand. In the long run, Taiwan is facing increasing competition from Korea in the high-end supply, and from mainland China in the medium- to low-end supply of the semiconductor market. The latest rebound in Taiwanese share prices is unsustainable, and they are about to relapse anew. Within an EM equity portfolio, we recommend staying neutral on Taiwanese stocks for now. Feature Taiwan’s exports and manufacturing are in full-blown recession. The equity market has rebounded after a major selloff last year. However, the overall manufacturing PMI and its export sub-component are extremely weak, and do not justify the latest share-price rebound (Chart I-1). Chart I-1Taiwanese Equities: Unsustainable Rally Taiwanese Equities: Unsustainable Rally Taiwanese Equities: Unsustainable Rally Are manufacturing activity and exports about to recover? Or will the stock market rally fade? Our answer is the latter. There are currently no signs suggesting a recovery in exports is imminent. Moreover, the engine of the economy – the semiconductor sector – is facing both cyclical and structural headwinds. We remain negative on Taiwanese stocks in absolute terms. Within an EM equity portfolio, we recommend a market-weight allocation to Taiwanese stocks for now. Importance Of Semiconductors Over the past 15 years, the semiconductor sector has become the cornerstone of the Taiwanese economy. The Taiwanese economy is highly dependent on its external sector, as exports contribute to nearly 70% of GDP. As such, Taiwan’s business cycle has often been closely associated with its export sector. This means the region’s growth outlook relies on both external demand (a cyclical factor) and the competitiveness of its export sector (more of a structural factor). Over the past 15 years, the semiconductor sector has become the cornerstone of the Taiwanese economy. It contributes to over one-third of the region’s total exports, up from 22% in 2009 (Chart I-2). Chart I-2Semiconductor: Cornerstone Of Taiwanese Economy Semiconductor: Cornerstone of Taiwanese Economy Semiconductor: Cornerstone of Taiwanese Economy Consistently, tech stocks also account for the lion’s share of the Taiwanese stock market, representing nearly 60% of the MSCI Taiwan Index and 47% of the Taiwanese Stock Exchange (TSE) index in market-value terms. There have been two key forces behind the significant growth of Taiwan’s semiconductor sector: booming global demand for smartphones/tablets and increasing competitiveness among domestic semiconductor companies. However, looking forward, the Taiwanese manufacturing sector and its semiconductor exports are facing a double-whammy: cyclical weakness in global demand and a relative decline in Taiwan’s export ability. In the context of a negative structural outlook, a cyclical downtrend engenders substantial deterioration in manufacturing, and by extension corporate profitability. Cyclical Downturn In Global Semiconductor Demand The outlook for the Taiwanese semiconductor industry remains poor. The global semiconductor industry has already been in a cyclical downtrend since early 2018. Global smartphone sales are shrinking. Both DRAM and NAND prices have been falling (Chart I-3). Chart I-3Falling Memory Chips Prices Falling Memory Chips Prices Falling Memory Chips Prices The freefall in Taiwan's new export orders seems to entail a further contraction in exports (Chart I-4). Chart I-4A Further Contraction In Exports Is Likely A Further Contraction In Exports Is Likely A Further Contraction In Exports Is Likely Importantly, exports of electronics parts lead Taiwanese tech EPS growth, and currently point to an impending contraction in corporate earnings (Chart I-5). Chart I-5An Impending Contraction In Corporate Earnings An Impending Contraction In Corporate Earnings An Impending Contraction In Corporate Earnings The outlook for the Taiwanese semiconductor industry remains poor. First, Taiwanese semiconductor producers are highly vulnerable to any further downside in global smartphone demand. There are two major pure-play wafer manufacturers in Taiwan: Taiwan Semiconductor Manufacturing Company (TSMC) and United Microelectronics (UMC). TSMC and UMC are the world’s largest and fourth-largest dedicated integrated circuit (IC) foundries, respectively. The smartphone sector has been the main revenue source for both companies, accounting for a 45% share for TSMC and 40% for UMC. Global smartphone demand is likely to decline further in 2019, as major markets such as mainland China and advanced economies have entered the saturation phase of mobile-phone demand. DRAMeXchange expects global smartphone production volume for 2019 to fall by 3.3% from last year following a 4% drop in 2018 (Chart I-6). Chart I-6Global Smartphone Demand Started A Downtrend Global Smartphone Demand Started A Downtrend Global Smartphone Demand Started A Downtrend Smartphone sales in mainland China remain in deep contraction after two consecutive years of declines (Chart I-7). Odds are that smartphone shipments will remain sluggish amid the ongoing economic slump in the mainland’s economy. Chart I-7Smartphone Sales In Mainland China Are In A Deep Contraction Smartphone Sales In Mainland China Are In A Deep Contraction Smartphone Sales In Mainland China Are In A Deep Contraction In addition, Taiwan’s TSMC is the sole chip supplier for Apple iPhones. A further decline in Apple smartphone shipments will reduce the company’s revenue and profits, damaging the region’s growth outlook. Mainland China now can produce top-notch quality smartphones at relatively cheaper selling prices. This will further crowd out higher-priced products from Apple, Samsung and others (Chart I-8). Chart I-8Apple Has Been Losing Market Share In Global Smartphone Market Apple Has Been Losing Market Share In Global Smartphone Market Apple Has Been Losing Market Share In Global Smartphone Market Second, the significant surge in bitcoin prices greatly boosted cryptocurrency mining activity in 2016-‘17 as miners quickly expanded their computing power. This boosted demand for graphic process unit (GPU) chips and in turn brought higher revenue for Taiwan chipmakers between June 2016 and early 2018. However, with the bust in bitcoin prices (Chart I-3 on page 3), demand from cryptocurrency mining has vanished and is unlikely to revive soon. Indeed, Taiwan chipmakers have suffered from last year’s plunge in cryptocurrency mining activity. According to TSMC, revenue from the cryptocurrency mining-related high-performance computing (HPC) sector contracted by double digits in 2018. Given that HPC demand is the second-biggest source of revenue for TSMC, with 32% share, TSMC revenue will be curtailed as HPC chip demand will continue to decline on weak bitcoin prices. Last, developments in new technologies, such as foldable smartphones, artificial intelligence, fifth-generation (5G) mobile networks and the so-called Internet of Things (IoTs) could only produce a modest pick-up in semiconductor demand. Most of these developments are still in their infancy and early stages. Hence, their growth will not be large enough to make a cyclical difference in global semiconductor demand. For example, the foldable smartphone that Huawei recently announced is indeed appealing. However, a lack of stability in panel supply and quite-high selling prices will limit sales. WitsView, a division of TrendForce, predicts that the market penetration rate of the foldable phone will be only 0.1% in 2019, and could rise to 1% in 2020 if more panel providers join the game, enabling a significant reduction in panel costs. Moreover, these categories together account for only ~23% of TSMC’s revenue; their modest growth will not be able to make up for the losses from the smartphone and HPC sectors within Taiwan’s economy. Besides, there has been a slowdown in demand from high-growth areas such as data center servers, as well as the automotive and industrial sectors. Putting it all together, odds are that global semiconductor demand will only materially recover in 2020. By that time, more-mature 5G technology and the increasing adoption of the 5G network and 5G-related products may be able to shift global semiconductor demand from the current downturn to a cyclical uptrend. Hence, the cyclical weakness in global semiconductor demand is likely to persist over the next six months. Consequently, Taiwan’s major types of semiconductor production will likely remain in contraction, and inventory levels will stay elevated (Chart I-9 and Chart I-10). Chart I-9Taiwan: Semiconductor Output Contraction Will Likely Continue Taiwan: Semiconductor Output Contraction Will Likely Continue Taiwan: Semiconductor Output Contraction Will Likely Continue Chart I-10Taiwan: Semiconductor Inventory Are Elevated Taiwan: Semiconductor Inventory Are Elevated Taiwan: Semiconductor Inventory Are Elevated Bottom Line: There are no signs of an imminent recovery in exports. A Potential Decline In Taiwan’s Semiconductor Competitiveness Taiwan wafer manufacturers are facing an increasing threat from their Korean and mainland China competitors. Leadership in advanced process technologies has been a key factor in Taiwan’s strong market position in the global semiconductor industry. With cutting-edge technologies, Taiwan has been the global wafer capacity leader since 2015. As of last year, it held about 22% of global installed wafer capacity (Chart I-11). Chart I-11 However, Taiwan wafer manufacturers are facing an increasing threat from their Korean and mainland China competitors. Korean Chipmakers While Taiwan will remain highly competitive in 7 nanometer (nm) and 10 nm wafer production, it is facing fierce competition from Korea. Manufacturing technologies designated by smaller nanometer numbers tend to have faster speeds and be more power-efficient than technologies designated by larger numbers. TSMC was the first company in the world to mass-produce 7 nm node wafers. Its 7 nm deep ultraviolet lithography (DUV) node has been in mass production since April 2018, producing chips for AMD, Apple, HiSilicon, and Xilinx. Beginning at the end of this month, TSMC will be ready to begin mass production of 7nm wafers using extreme ultraviolet lithography (EUV). The switch from 7nm DUV to 7nm EUV allows for fewer defects and fewer steps required during the production process. The company also aims to boost volume production of its 5 nm nodes in early 2020 and has a target of 3 nm wafers for 2022. Last year, wafer revenue from 7nm and 10nm chips accounted for 9% and 11% of TSMC’s total revenue, respectively (Chart I-12). Chart I-12 Samsung has been closely following TSMC in terms of technological innovation. It started mass production of EUV-based 7nm chips last October, with a plan of risk production1 of 5nm wafers in 2019 and a target of 4nm wafers in 2022. Meanwhile, IBM announced last December that it signed an agreement with Samsung to produce its next-generation processors with Samsung’s 7nm technology. As Samsung seeks to diversify its revenue source away from memory chips, which last year contributed to about 80% of its operating profit, the company has been determined to ramp up the development of its foundry business. It aims to replace TSMC as the world’s largest foundry producer by 2030. In the near term, Samsung aims to secure a 25% market share in the global pure-play foundry market by 2023, a rise from 19% currently. Last year, Samsung surpassed Taiwan’s UMC to become the world’s second-largest dedicated chipmaker. Moreover, Samsung’s capital spending has been and will continue to be much higher than TSMC. Over the course of 2017 and 2018, Samsung spent about $46.9 billion on semiconductor capital expenditures, more than double TSMC’s $21 billion. Hence, the competition between TSMC and Samsung in the high-end chip market will intensify in the coming years. Chipmakers In Mainland China The competition between TSMC and chipmakers from mainland China is also escalating. Chart I-12 shows that 80% of TSMC’s wafer revenue comes from bigger node wafers (bigger than 10 nm). Taiwan’s second-biggest chipmaker, UMC, only produces wafers equal to or bigger than 28 nm. Therefore, the chip market using less-advanced technology than 10 nm will be the main battlefield between Taiwanese and mainland China’s chipmakers. Before 2014, there were few wafer manufacturers in mainland China, and those that did exist were too weak to compete with giant market players like TSMC. In 2014, the Chinese central government made a move to foster development within the local IC industry. Since then, the authorities have poured significant amounts of capital into semiconductor foundries, as well as companies focused on memory production, chip design and related equipment and materials. Semiconductor Manufacturing International Corporation (SMIC) is the world’s fourth-largest dedicated wafer manufacturer, and is the largest in mainland China. While 28nm will likely remain a large part of its business, SMIC plans to go into production on its 14 nm technology in the first half of 2019. The company is also working on 10nm and 7nm nodes with the use of EUV. SMIC currently counts HiSilicon and Qualcomm as customers, manufacturing smartphone chips with medium-to-low technology. As mainland China aims to increase its self-sufficiency rate for ICs significantly over the next five to 10 years, the nation’s producers will significantly expand their wafer capacity. Mainland China is likely to reduce its semiconductor imports from Taiwan considerably in the coming years, especially wafer imports. According to IC Insights, nine 300mm wafer fabs2 are scheduled to open worldwide in 2019, with five of them in mainland China. Based on another set of data from SEMI, the number of 200mm wafer fabs in the world will increase from 194 in 2017 to 203 by 2022, with an additional 56 established fabs planning to expand their manufacturing capacity. Mainland China is expected to account for 44% of the growth. In comparison, Taiwan only accounts for about 10% of the growth. Mainland China currently accounts for over 30% of Taiwanese electronic parts exports (wafers, PCBs, mainboards and others). As mainland China continues to build new wafer manufacturing capacity and gradually improve its existing technology, it will switch its consumption from imports to domestic production. Consequently, mainland China is likely to reduce its semiconductor imports from Taiwan considerably in the coming years, especially wafer imports (Chart I-13). This is structurally bearish for Taiwanese semiconductor companies. Chart I-13Mainland China’s Semiconductor Imports From Taiwan Will Drop Mainland China’s Semiconductor Imports From Taiwan Will Drop Mainland China’s Semiconductor Imports From Taiwan Will Drop Bottom Line: Taiwan is facing increasing challenges from Korea in terms of defending its market share in the high-end wafer market. Meanwhile, Taiwan is also set to lose market share in the medium-to-low market to wafer producers from mainland China. What About The Rest Of The Economy? The rest of the economy is exhibiting mixed signals, with contracting major non-semiconductor export sectors but decent household consumption and property market. Table 1 shows Taiwan’s top 10 exported products, with the top three attributing to over half of total exports. Besides the semiconductor sector, exports of the other two major products – electrical machinery products and machinery – are beginning to contract (Chart I-14). Chart I- Chart I-14Taiwan: Contracting Non-Semiconductor Exports Taiwan: Contracting Non-Semiconductor Exports Taiwan: Contracting Non-Semiconductor Exports However, the domestic economy seems to be running well at present. Production of construction materials in volume terms is growing rapidly, accompanied by a rebound in building permits granted (Chart I-15). While employment growth is decent, average wage growth has been quite strong (Chart I-16). With persistent contraction in exports and inflation very low, the central bank could cut rates in 2019. Chart I-15Decent Domestic Demand Decent Domestic Demand Decent Domestic Demand Chart I-16Strong Wage Growth Strong Wage Growth Strong Wage Growth Ongoing contraction in semiconductor exports will likely slow domestic demand with a time lag. In fact, the inverted 5-year/6-month yield curve is indeed signaling an economic slump in Taiwan (Chart I-17). Chart I-17Inverted Yield Curve Signals Continuing Economic Slump Ahead Inverted Yield Curve Signals Continuing Economic Slump Ahead Inverted Yield Curve Signals Continuing Economic Slump Ahead Investment Recommendations The latest rebound in Taiwanese stocks is unsustainable and share prices will relapse again. Within an EM equity portfolio, we recommend maintaining a market-weight allocation to Taiwan for now. We are reluctant to downgrade Taiwan to underweight because some other emerging markets and sectors within the EM universe have a poorer outlook. In addition, Taiwanese shares have already underperformed the EM benchmark since last September (Chart I-18). Chart I-18Taiwanese Stocks: Staying Neutral Within EM Taiwanese Stocks: Staying Neutral Within EM Taiwanese Stocks: Staying Neutral Within EM The Taiwanese currency is cheap (Chart I-19). The region has a massive current account surplus and foreigners do not hold any local bonds, which is very different from many other EM countries. Hence, Taiwan is less vulnerable to capital outflows than many current-account-deficit EM economies. The latter could be forced to raise rates, which will place pressure on their banks as well as on domestic demand. In contrast, Taiwan has the ability to cut rates. Chart I-19TWD Is Cheap TWD Is Cheap TWD Is Cheap Ellen JingYuan He, Associate Vice President Emerging Markets Strategy ellenj@bcaresearch.com     1 "Risk Production" means that a particular silicon wafer fabrication process has established a baseline in terms of process recipes, device models, and design kits, and has passed standard wafer level reliability tests. 2 A fab, sometimes called foundry, is a semiconductor fabrication plant where devices such as integrated circuits are manufactured. Equity Recommendations Fixed-Income, Credit And Currency Recommendations

Highlights Korean stocks are facing downside risks over the next several months. Exports will continue to contract on falling semiconductor prices and retrenching global demand. Growth deceleration and low inflation will lead the central bank to cut rates in 2019. Within an EM equity portfolio, we are downgrading Korean tech stocks from overweight to neutral but remain overweight the non-tech sector. We are booking gains on our strategic long positions in EM tech versus both the broader EM equity benchmark and materials. The KRW/USD exchange rate is at a critical technical juncture. Investors should wait to buy on a breakout and/or sell on a breakdown of the tapering wedge pattern. Feature   Decelerating and lately contracting South Korean exports have been a major drag on the economy and stock market (Chart I-1). The country is heavily reliant on manufacturing, with exports of goods contributing to nearly half of real GDP. Chart I-1Korean Stocks: Unsustainable Rebound? Korean Stocks: Unsustainable Rebound? Korean Stocks: Unsustainable Rebound? Although exports are currently shrinking, Korean domestic stock prices still rebounded. The rebound has mostly been driven by the information technology (tech) sector (Chart I-2). Chart I-2 Is this recent rally justified by underlying fundamentals? Will share prices continue to rise in 2019? Our inclination is ‘no’ to both questions. There are still dark clouds on the horizon for both Korea’s business cycle and stock market. We are downgrading Korean tech stocks to neutral from overweight within a dedicated EM equity portfolio. However, we are maintaining our overweight in non-tech stocks relative to the EM equity benchmark. Lingering Risks In The Semiconductor Industry Korea’s dependence on the semiconductor sector has risen considerably in the past several years: Semiconductor exports have risen from under 10% to slightly above 20% of total goods exports (Chart I-3). As such, the outlook for semiconductor exports is a critical factor for future economic growth. Chart I-3Korea: Increasing Reliance On The Semiconductor Sector Korea: Increasing Reliance On The Semiconductor Sector Korea: Increasing Reliance On The Semiconductor Sector Table 1 lists the top 10 major exported goods from Korea, together contributing about 72% of total exports. Semiconductors are by far the largest component. Last year, overseas sales of semiconductors alone contributed to some 90% of growth in Korean exports, and about one-third of the country’s nominal GDP growth. Chart I- Notably, Korea produces the largest quantity of DRAM and NAND memory chips in the world. Last year, Korean semiconductor companies accounted for about 70% of global DRAM and 50% of NAND flash global sales revenue. In 2019 Korean semiconductor exports will likely contract due to further deflation in DRAM and NAND memory prices (Chart I-4). Chart I-4Memory Prices Are Plunging Memory Prices Are Plunging Memory Prices Are Plunging The 2016-2017 surge in DRAM and NAND flash prices was due to supply shortages relative to demand. Last year, NAND prices plunged and DRAM prices began to fall as their supply-demand balances shifted to oversupply. This year, the glut will worsen. Demand Global demand for DRAM and NAND memory is slowing. Memory demand from the global smartphone sector – one important end-user market for DRAM and NAND memory chips – is contracting. According to the International Data Corporation (IDC), the global mobile phone sector is the biggest end-market for both DRAM and NAND memory chips, with nearly 40% market share in each. As major markets like China and advanced economies have entered the saturation phase of mobile-phone demand, global smartphone shipments are likely to decline further in 2019 (Chart I-5, top panel). Chart I-5Global Memory Demand Is Slowing Global Memory Demand Is Slowing Global Memory Demand Is Slowing DRAMeXchange1 expects global smartphone production volume for 2019 to fall by 3.3% from last year. In addition, the significant surge in bitcoin prices greatly boosted cryptocurrency mining activity in 2016-‘17 as miners quickly expanded their computing power. This contributed to strong DRAM demand and in turn higher semiconductor prices between June 2016 and May 2018. With the bust of bitcoin prices, this demand has vanished, which will further weigh on prices (Chart I-5, bottom panel). Supply High semiconductor prices in 2016-2017 boosted global production capacity expansion of DRAM and NAND memory chips. Based on data compiled by the IDC, global DRAM and NAND flash capacity expanded by 5.7% and 4.3% respectively in 2018 from a year earlier. As most of the global new capacity was added in the second half of 2018, the output of DRAM and NAND in 2019 will be higher than last year. Moreover, DRAM capacity will grow an additional 4% this year. Because of rising supply and slowing demand, both DRAM and NAND markets are in excess supply and have high inventories. DRAMeXchange forecasts that average DRAM prices will drop by at least another 20% in 2019, while NAND flash prices will fall another 10% from current levels. DRAM and NAND flash memory are the largest components of Korean tech producers. Yet they also sell many other tech products such as analog integrated circuits, LCD drivers, discrete circuits, sensors, actuators, and so on. Apart from the negative impact of declining global DRAM and NAND flash prices, the country’s semiconductor exports will also suffer from slowing demand in China in 2019. China, the biggest importer of Korean semiconductor products, has already shown waning demand. Its imports of electronic integrated circuits and micro-assemblies have contracted over the past two months in both value and volume terms (Chart I-6, top and middle panels). This mirrors a similar contraction in Korean semiconductor exports over the same period (Chart I-6, bottom panel). Chart I-6Weakening Chinese Semiconductor Demand Weakening Chinese Semiconductor Demand Weakening Chinese Semiconductor Demand Bottom Line: Korean semiconductor producers will likely face a contraction in their sales in 2019 due to weakening demand and deflating semiconductor prices. Diminishing Competitive Advantage Korea has been losing its competitive edge in key sectors like automobiles and smartphones. Even though the country remains highly competitive in the global semiconductor industry, it is beginning to show early signs of losing competitiveness there too. Improving competitiveness among other producers as well as a slowing pace of technological improvement and rising production costs are major reasons underlying Korea’s diminishing global competitiveness. Automobiles Korean auto manufacturers have lost market share in the global auto market. In China, the world’s biggest auto market, Korean brands’ market share has declined significantly in the past four years, losing out to both Japanese and German brands (Chart I-7, top three panels). Chart I-7Korea: Losing Market Shares In China's Auto Market Korea: Losing Market Shares In China's Auto Market Korea: Losing Market Shares In China's Auto Market Korean car companies have established auto manufacturing plants in China over the past decade. As a result, all Korean cars sold in China are produced within China, and automobile exports to China from Korea have fallen to zero (Chart I-7, bottom panel). Due to Korean auto manufacturers’ diminishing competitive advantage, Korean automobile production and exports peaked in 2012 in terms of volumes, and have been on a downtrend over the past seven years (Chart I-8, top panel). Chart I-8Further Decline In Korean Auto Output And Exports Is Possible Further Decline In Korean Auto Output And Exports Is Possible Further Decline In Korean Auto Output And Exports Is Possible While demand for Korean cars in the EU remains resilient, sales volumes in the U.S., China and the rest of world have been on a downward trajectory (Chart I-8, bottom three panels). Smartphones In the global smartphone market, Korea’s major smartphone-producing company – Samsung – has been in fierce competition with Chinese brands, and it seems to be losing the battle. Chart I-9 shows that while Samsung’s smartphone sales declined 8% year-on-year last year, smartphone sales from major Chinese smartphone producers (Huawei, Xiaomi, Oppo and Vivo) continued to grow at a pace of 20%. Chart I-9Korea: Losing Market Shares In Global Smartphone Market Korea: Losing Market Shares In Global Smartphone Market Korea: Losing Market Shares In Global Smartphone Market From 2012 to 2018, China’s share of global smartphone shipments rose from 6% to 39%. By comparison, Samsung’s share declined from 30% to 21% over the same period. Semiconductors Korean semiconductor companies – notably Samsung and SK Hynix – will likely remain the biggest producers in the memory market, given their advanced technology. However, there are still signs that Korean semiconductor companies will face increasing challenges in protecting their market share. Based on IDC data, Korean semiconductor companies’ share of global DRAM capacity will inch lower to 65% in 2019 from 65.4% in 2017, while their share of NAND capacity will decline to 53.8% from 57.5% during the same period. Meanwhile, China is focusing on boosting its self-sufficiency in terms of semiconductor production. At the moment there is still a three- to four-year technological gap between China and Korea in DRAM and NAND mass production, though the gap is likely to narrow. In the meantime, the U.S. will continue to create obstacles to prevent the rise of the Chinese semiconductor sector. However, these factors will only delay – not avert – the sector’s development and growth. We believe China will remain firmly committed to develop its semiconductor sector, particularly memory products, irrespective of the cost of investment necessary to do so. Similar to what has transpired in both automobile and smartphone production (Chart I-10), China will slowly increase its penetration in the semiconductor market with increasing capacity and a narrower technology gap over the next five to 10 years. After all, the world’s biggest semiconductor demand is in China. Chart I-10China: A Rising Star In Global Auto And Smartphone Market China: A Rising Star In Global Auto And Smartphone Market China: A Rising Star In Global Auto And Smartphone Market Significant increase in labor costs = falling export competitiveness for all sectors Korean President Moon Jae-in’s flagship economic policy, “income-led growth,” has resulted in dramatic increases in minimum wages since he took office in 2017, further damaging Korea’s competitiveness. The nation’s minimum wage was hiked by 7.3% in 2017, 16.4% in 2018 and will rise by another 11% to 8,350 KRW or $7.40 an hour, in 2019. As the president remains committed to meeting his campaign pledge of lifting the minimum wage to 10,000 KRW an hour, or about $8.90, this would require a further 20% increase in the next year or two. In addition, the government has also limited the maximum workweek to 52 hours since last July for businesses with more than 300 workers. Last month, the Cabinet further approved a revision bill whereby workers are eligible to receive an additional eight hours of wages every weekend for 40 hours of work that week. The new wage regulations have become a substantial burden on employers in all industries. The impact is more severe on small- and medium-sized enterprises (SMEs). According a recent survey, about 30% of SMEs have been unable to pay workers due to the state-set minimum wage. It is also affecting large manufacturers. According to a joint statement released in late December by the Korea Automobile Manufacturers Association and the Korea Auto Industries Cooperative Association, local automakers’ annual labor cost burdens will increase by at least 700 billion won (US$630 million) a year. As for auto parts manufacturers, a skyrocketing financial burden due to the new policy may threaten their survival. In addition, despite the KORUS FTA agreement reached between Korea and the U.S. last September, Korean auto manufacturers still fear they will be subject to new tariffs in 2019. On February 17, the U.S. Commerce Department submitted a report about imposing tariffs on imported automobiles and auto parts to U.S. President Donald Trump, who will make a decision by May 18. Our Geopolitical Strategy Service (GPS) team believes the odds of U.S. administration imposing auto tariffs on imported cars from Korea are small as this will be against the KORUS FTA agreement.2 Our GPS team also believes Japan is less likely to suffer a tariff than the EU, and even if Japan suffers a tariff along with the EU, Japan will negotiate a waiver more quickly than the EU. In both cases, Korea is likely to sell more cars in the U.S., but it will continue to face strong competition from Japan. Bottom Line: In addition to weakening global demand, a deterioration in Korea’s competitive advantage, due in large part to improving competitiveness among other producers and rising domestic wages, will negatively affect Korean exports. What About Domestic Demand? Record fiscal spending in 2019 will boost public sector consumption considerably, offsetting weakening consumption in the private sector. As the new wage policy will likely result in more layoffs and additional shuttering of businesses, domestic retail sales growth will remain under pressure (Chart I-11). Hence, an unintended consequence of the government’s higher income policy will be weaker aggregate income and consumer spending growth. Chart I-11KOREA The New Wage Policy May Trigger More Layoffs And Weaken Retail Sales KOREA The New Wage Policy May Trigger More Layoffs And Weaken Retail Sales KOREA The New Wage Policy May Trigger More Layoffs And Weaken Retail Sales Manufacturing and service sector jobs, including wholesale and retail trade and hotels and restaurants, account for 17% and 23% of total employment, respectively. Of all sectors, these two lost the most employees in January from a year ago. Meanwhile, due to the government’s deregulation of loans in 2014, Korean household debt has increased at a much faster pace than nominal income growth (Chart 12, top panel). As a result, Korea’s household debt has rapidly risen to 86% of its GDP as of the end of the third quarter of last year, from 72% four years ago – (Chart I-12, bottom panel). Elevated household debt at a time of rising layoffs will increase consumer anxiety and weigh on household spending. Chart I-12High Household Debt Will Weigh On Spending High Household Debt Will Weigh On Spending High Household Debt Will Weigh On Spending In order to combat an economic downturn, the government last month approved a record 467 trillion won ($418 billion, 26.5% of the country’s 2018 GDP) budget for 2019, up 9.5% from last year. The last time the budget increased by such a big scale was in 2009, when spending rose 10.7% in the wake of the global financial crisis. In addition, the government will front-load spending – with 61% of the budget to be spent in the first half of 2019. Household spending and government expenditures account for 48% and 15% of real GDP, respectively, while exports equal about 50% of real GDP. Hence, the increase in fiscal spending will not entirely offset the contraction in exports and slowdown in consumer spending. This entails a considerable slowdown in economic growth in 2019. Bet On Monetary Easing With growth disappointing and both headline and core inflation well below 2% (Chart I-13), the central bank will cut rates in 2019. Chart I-13Bet On A Rate Cut Bet On A Rate Cut Bet On A Rate Cut So far, economic growth has decelerated in the past 10 months, and recent data shows no signs of recovery. The country’s manufacturing sector is in contraction, with manufacturing PMI holding below the 50 boom-bust line in January (Chart I-14). Meanwhile, South Korea's unemployment rate rose to a nine-year high in January, with most of the job losses in the manufacturing and construction sectors. Chart I-14Manufacturing Sector: Still In Contraction Manufacturing Sector: Still In Contraction Manufacturing Sector: Still In Contraction Saramin, a South Korean job search portal, surveyed 906 firms in South Korea last month, 77% of which expressed unwillingness to hire new employees due to higher labor costs and negative business sentiment. Retail sales volume growth recently tumbled to 2-3%, pointing to faltering domestic demand (Chart I-11 above, bottom panel). The fixed-income market is not pricing in a rate cut in 2019. Therefore, investors should consider betting on lower interest rates. Shrinking exports and rate cuts will likely undermine the Korean won. Bottom Line: Economic deceleration and low inflation will lead the central bank to cut interest rates in 2019. Investment Implications The following are our investment recommendations: Downgrade the Korean tech sector from overweight to neutral within the EM space. We are reluctant to downgrade to underweight because many other emerging markets and sectors within the EM universe have poorer structural fundamentals than Korean tech. The tech sector accounts for 38% of the MSCI Korea Index, and 27% of the KOSPI in terms of market value. The stock with the largest weight in the MSCI Korea equity index is Samsung Electronics, with a share of 25%, followed by SK Hynix, with a ~5% share. Both are very sensitive to semiconductor prices. Specifically, semiconductor sales accounted for 31% of Samsung’s revenue, but contributed 77% of Samsung’s operating profit last year (Table I-2). Chart I- Falling prices reduce producers’ profits by more than falling volumes.3 Hence, profits of semiconductor producers in Korea and globally will shrink in 2019. This will lead to a substantial selloff in Korean tech stocks (Chart I-15). Chart I-15Falling Memory Prices Will Trigger A Sell-Off In Korean Tech Stocks Falling Memory Prices Will Trigger A Sell-Off In Korean Tech Stocks Falling Memory Prices Will Trigger A Sell-Off In Korean Tech Stocks Meanwhile, China accounts for 33% of Samsung’s revenue, making it the largest market (Chart I-16). The ongoing economic slump in China’s domestic demand implies weaker demand for Korean shipments to China, which account for 28% of its exports and 14% of its GDP. Chart I-16 ​​​​​​​ We are booking gains on our strategic long position in the Korean tech sector versus the EM benchmark index first instituted on January 27, 2010. This trade resulted in a 136% gain (Chart I-17, top panel). Chart I-16Taking Profits On Our Overweight Tech Positions Taking Profits On Our Overweight Tech Positions Taking Profits On Our Overweight Tech Positions Consistently, we are also taking profits on our long EM tech / short EM materials stocks trade, a strategic recommendation initiated on February 23, 2010 that has yielded a 186% gain (Chart I-17, second panel). The basis for this strategic position 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, 2010.4 Stay overweight non-tech equities within the EM space. The fiscal stimulus will have a considerable positive impact on the economy. Besides, Korean non-tech stocks have been weak relative to the EM equity benchmark, and in a renewed EM selloff they could act as a low-beta play (Chart I-17, bottom panel). We initiated our long Korean non-tech sector versus the EM benchmark index on May 31, 2018, which has so far been flat. The KRW/USD exchange rate is at a critical technical juncture. Investors should wait and buy on a breakout or sell on a breakdown of the tapering wedge pattern. The KRW/USD has been in a tight trading range over the past eight months (Chart I-18) and is approaching a major breaking point – i.e., any move will be significant, which we expect will largely depend on the movement of the RMB/USD. Chart I-18Tapering Wedge Patterns Tapering Wedge Patterns Tapering Wedge Patterns The natural path for the RMB would have been depreciation versus the U.S. dollar. However, China may opt for a flat exchange rate versus the U.S. dollar given its promises to the U.S. within the framework of forthcoming trade agreements. We have been shorting the KRW versus an equally weighted basket of USD and yen since February 14, 2018. We continue to hold this trade for the time being. Investors should augment their positions if the KRW/USD breaks down or close this trade and go long the won if the KRW/USD breaks out of its tapering wedge pattern. With respect to fixed income, we continue to receive Korean 10-year swap rates as we expect interest rates to fall meaningfully. Local investors should overweight bonds versus stocks.   Ellen JingYuan He, Associate Vice President Emerging Markets Strategy ellenj@bcaresearch.com     Footnotes 1 DRAMeXchange, the memory and storage division of a technology research firm TrendForce, has been conducting research on DRAM and NAND Flash since its creation in 2000. 2 Please see the Geopolitical Strategy Weekly Report, "Trump's Demands On China", published April 4, 2018. Available at gps.bcaresearch.com. 3 Please see the Emerging Markets Strategy Weekly Report “Corporate Profits: Recession Is Bad, Deflation Is Worse”, dated January 28, 2016, available at www.bcaresearch.com 4 Please see the Emerging Markets Strategy Special Report “How To Play Emerging Market Growth In The Coming Decade”, dated June 8, 2010, available at www.bcaresearch.com. Equity Recommendations Fixed-Income, Credit And Currency Recommendations