## Annex Table 3.2. Asian Bond Markets: Second-Stage Results—Determinants of Beta Coefficients ## Source details **Canonical URL:** [Annex Table 3.2. Asian Bond Markets: Second-Stage Results—Determinants of Beta Coefficients](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16173.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16173.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16173.pdf.json) --- ### I. Main context and research questions - China is now the world’s second largest economy at market exchange rates. - During 2000–15, China accounted for nearly one-third of global growth. - Key questions: - What are the potential spillover channels from China to financial markets? - Are financial spillovers from China rising, particularly to Asian markets? - How to quantify the size of these spillovers and the roles of trade and financial linkages? ### II. First-stage (bond market) regression results — stylized findings - General finding: - Asian bond markets’ sensitivities to the Chinese bond market remain low except for a few exceptions (Korea and Taiwan Province of China). - Regional bond markets are more sensitive to the U.S. bond market and, to a lesser extent, to the Japanese bond market. - Factor loadings for the full sample (daily changes in 10-year government bond yields, percentage points; N and R2 shown where present): - Australia: China 0.003 / United States 0.522*** / Japan 0.531*** — N = 2,421 — R2 = 0.391 - India: China 0.053 / United States 0.089*** / Japan 0.087 — N = 2,421 — R2 = 0.016 - Indonesia: China 0.137 / United States -0.041 / Japan -0.087 — N = 2,421 — R2 = 0.009 - Korea: China 0.077** / United States 0.192*** / Japan 0.153*** — N = 2,421 — R2 = 0.090 - Malaysia: China 0.023 / United States 0.079*** / Japan 0.035 — N = 2,421 — R2 = 0.019 - New Zealand: China -0.005 / United States 0.347*** / Japan 0.269*** — N = 2,113 — R2 = 0.283 - Philippines: China 0.066 / United States -0.129*** / Japan 0.000 — N = 2,207 — R2 = 0.010 - Taiwan Province of China: China 0.045*** / United States 0.085*** / Japan 0.133*** — N = 2,421 — R2 = 0.115 - Thailand: China 0.089** / United States 0.146*** / Japan 0.093* — N = 2,421 — R2 = 0.041 - Significance notation preserved: "*** p<0.01, ** p<0.05, * p<0.1". ### III. Second-stage regression results — determinants of Asian bond market sensitivities - Main finding: - In contrast to equity market results, all three explanatory variables (trade linkages, trade competition, financial linkages) lose their degree of significance for bond markets. - Even with alternate specifications, results for Chinese bond market specifications remain insignificant. - Reported regression snippet (coefficients and robust t-statistics preserved exactly where presented): - Trade exposure: 0.004 0.128** (robust t-statistics in parentheses: (0.167) (2.521)) - Trade competition: -0.021 -0.000 (robust t-statistics: (-1.186) (-0.017)) - Financial linkages: 0.006 0.040 (robust t-statistics: (0.416) (0.827)) - Observations: 9090 - R-squared: 0.119 0.193 - Note: Robust t-statistics reported in parentheses. Significance notation preserved: "*** p<0.01, ** p<0.05, * p<0.1". ### IV. Interpretation relative to equity market results - Contrast with equity markets: - For equities, trade exposure is the primary transmission channel from China; financial linkages gained significance post-GFC and via Hong Kong SAR gateway. - For bonds, explanatory variables (trade linkages, trade competition, financial linkages) do not show consistent significance for explaining sensitivities to Chinese bond-market movements. - Implication: - Bond markets in the region appear primarily driven by U.S. and Japanese bond-market movements rather than by Chinese bond-market shocks, with Korea and Taiwan Province of China as notable exceptions. ### V. Data and methodological notes (as applied to bond results) - First-stage regression: daily changes in 10-year government bond yields (percentage points). - Sample sizes for first-stage factor loadings: N values reported per country (e.g., N = 2,421 for most series; N = 2,113 for New Zealand; N = 2,207 for Philippines). - Second-stage: pooled regressions with 9,090 observations reported; robust t-statistics used for inference. ### VI. Policy-relevant implications (from bond-market perspective) - Market focus: - Regional policymakers should note that bond-market spillovers from China are limited in the sample and specification used, relative to spillovers from the United States and Japan. - Risk monitoring: - Countries with elevated sensitivity (Korea and Taiwan Province of China) may require closer monitoring of cross-border bond-market exposures. - Broader context: - These bond-market results complement equity-market findings where trade linkages and post-GFC financial linkages (including via Hong Kong SAR) matter more for spillovers. *Source: Annex Table 3.2. Asian Bond Markets: Second-Stage Results—Determinants of Beta Coefficients (source document contents).* ### References ________________________________________________________________34 ### _wp16173 - References ________________________________________________________________34 ### Annexes - Annex 1. Event Study: Exceptionally Large Movements In Chinese Markets___________ 28 - Annex 2. Construction of The Dataset Used in the Empirical Analysis _________________30 - Annex 3. Empirical Results: Bond Markets ______________________________________33 ### Figures - Figure 1. China’s Contribution to Global Growth and Recent Market Turbulence _________4 - Figure 2. Asian Market Correlations with China and the United States _________________6 - Figure 3. Asia: Business Cycle Synchronization with China __________________________7 - Figure 4. Channels of Spillovers from a Slowdown in China _________________________8 - Figure 5. Asia: Trade Exposure to China _________________________________________9 - Figure 6. Asia: Financial Claims on China and Hong Kong SAR _____________________10 - Figure 7. Major Emerging Markets: Total External Liabilities _______________________11 - Figure 8. China: Outward Direct Investments and Bilateral Currency Swap Agreements, end-2015 _________________________________________________________12 - Figure 9. Selected Asia: Exchange Rate Depreciations during Risk-Off Episodes, 2010-16 __________________________________________________________13 - Figure 10. Regional Equity and Foreign Exchange Markets during China-Related Shocks _14 - Figure 11. Event Study: Asian Stock Market Movements when China Experiences Outsized Market Movements ________________________________________15 - Figure 12. Event Study: Asian Exchange Rate Movements when China Experiences Outsized Market Movements ________________________________________16 - Figure 13. Event Study: Asian Market Movements when China Experiences Outsized Market Movements, Conditional on the VIX ____________________________16 - Figure 14. Equity Market Spillovers ____________________________________________19 - Figure 15. Asian Stock Markets: Determinants of Beta Coefficients with Center Economies _______________________________________________________20 - Figure 16. Asian Stock Markets: Determinants of Beta Coefficients with China _________21 - Figure 17. Scenario Analysis: Transmission of Shocks _____________________________22 - Figure 18. Asian Market Sensitivity to China under Different Scenarios _______________22 ### Tables - Table 1. Global Risk-off Episode Since 1992 ____________________________________13 - Table 2. Asian Stock Markets: First-Stage Results––Estimated Beta Coefficients ________19 - Table 3. Asian Stock Markets: Second-Stage Results––Determinants of Beta Coefficients _20 - Table 4. Asian Stock Markets: Financial Linkages with China: Pre-GFC vs. Post-GFC ___21 - Table 5. Sensitivity Tests: Drivers of Asian Equity Market Sensitivity to China _________24 ### Annex Tables - Annex Table 1.1. Event Study: Exceptionally Large Movements in the Chinese Market ____________________________________________________28 - Annex Table 1.2. Event Study: Exceptionally Large Movements in the Chinese Onshore Spot Exchange Rate (CNY) ____________________________________29 - Annex Table 2.1. Correlation between the Variables in First Stage ____________________32 - Annex Table 2.2. Correlation between the Variables in Second Stage _________________32 - Annex Table 3.1. Asian Bond Markets: First Stage Results––Estimated Beta Coefficients _33 *Source: _wp16173 - References ________________________________________________________________34* ### Annex Table 3.2. Asian Bond Markets: Second-Stage Results––Determinants of Beta ### Annex Table 3.2. Asian Bond Markets: Second-Stage Results––Determinants of Beta ### I. Introduction — Context and research questions - China is now the world’s second largest economy at market exchange rates. - During 2000–15, China accounted for nearly one-third of global growth. - Exports to China increased from 3 percent to 9 percent of world exports and from 9 percent to 22 percent of Asian exports over the same period. - China’s total external liabilities are around US$5 trillion, and China’s external assets, including reserves, are US$6.2 trillion as of 2015. - Key questions addressed: - What are the potential spillover channels from China to financial markets? - Are financial spillovers from China rising, particularly to Asian markets? - How to quantify the size of these spillovers and the roles of trade and financial linkages? ### Major empirical findings and robustness - Main findings: - Financial spillovers from China to regional markets are on the rise, especially in equity and foreign exchange markets. - Spillovers are stronger for economies with greater trade linkages with China. - The role of financial linkages in driving spillovers is growing in importance. - Empirical analysis using daily data from 2001 to 2014 shows major Asian equity markets are increasingly influenced by China's equity market, although not yet to the level of the United States. - China-related shocks coupled with global risk aversion episodes (e.g., August 2015 and January 2016) can produce much larger impacts on regional markets. - Robustness: - Results are virtually identical whether stock returns are measured in dollars or local currency, or whether they are expressed in absolute or excess returns. - Main results hold broadly for commodity exporters and importers. - Controlling for macro-financial vulnerabilities or the presence of common investors does not change the main findings. ### III. Co-movements in Asian markets — Stylized facts - Correlations: - Average correlation of returns in Asian markets with China has risen over time (equity and foreign exchange). - Asia’s bond markets have remained uncorrelated with the Chinese bond market. - Asia’s equity and bond market correlations with the United States have remained high. - Business cycle synchronization: - Countries with higher business cycle synchronization with China tend to have equity markets that move more closely with China. ### IV. Channels of financial spillovers from China - Four distinct transmission channels for negative China news to other countries’ asset markets: - Direct trade linkages: slowdown in China lowers exports of trading partners, weakening stock valuations and local currencies. - Indirect trade (commodity price) linkages: weaker China reduces global commodity prices, affecting commodity exporters’ valuations and currencies. - Direct financial linkages: losses on international investor positions in China affect other markets. - Indirect financial linkages: global risk premiums rise in response to China shocks, affecting other countries. - Heterogeneity: - Exposure varies by country across channels. ### V. Trade linkages — Direct and indirect effects - Direct trade linkages: - China accounts for about 50 percent of intraregional trade flows of imported inputs, doubling since 1995. - Value added in exports embedded into final demand in China exceeds 4 percent of GDP for Australia, Korea, Malaysia, Singapore, Taiwan Province of China, Thailand, and Vietnam. - A 10 percent decline in Chinese import volumes could subtract 0.4 percentage points or more off growth from these economies. - Indirect trade linkages: - Trade competition with China can generate positive or negative spillovers depending on the nature of the China-specific shock (e.g., competitiveness vs. exchange rate depreciation). - China’s slowdown is associated with lower metal prices; impact on oil and food prices appears more limited. - Value-added trade measures used capture exposures through third countries. ### VI. Financial linkages — Direct and indirect financial transmission - Direct financial linkages: - Financial claims on China and Hong Kong SAR combined were more than 10 percent of GDP for Korea, Singapore, and Taiwan Province of China as of end-2014. - Banks in the Asia-Pacific region accumulated about US$1.2 trillion of China-related exposures by the end of 2015. - At end-2015 cross-border loans to China accounted for 27 percent of banking system assets in Hong Kong SAR, 15 percent in Singapore, and 7 percent in Taiwan Province of China. - Foreign claims on China approach US$5 trillion. - In 2015, almost all of the capital outflows from emerging markets were accounted for by China. - China’s total external assets as of end-2015: US$6.2 trillion consisting of reserves US$3.4 trillion, portfolio investment US$261 billion, outward direct investment US$1.1 trillion, and other investments US$1.4 trillion. - China’s outward direct investments exceeded US$1 trillion at end-2015, with an estimated US$300 billion to Asia, representing 6.5 percent of recipient-country GDP on average. - Chinese banks’ overseas loans increased by more than US$600 billion since 2010 to near US$1 trillion at end-2015. - China launched more than 30 bilateral currency swap agreements since 2008, with an outstanding amount of US$500 billion at end-2015; more than half with Asian economies. - Potential future adjustments: Bayoumi and Ohnsorge (2013) estimate capital account liberalization could lead to a stock adjustment of Chinese assets abroad on the order of 15–25 percent of GDP and foreign assets in China on the order of 2–10 percent of GDP. - Inclusion of China’s domestic A-shares in global indices (e.g., MSCI Emerging Markets Index followed by an estimated US$1.5 trillion of funds) could trigger large portfolio rebalancing. - Indirect financial linkages: - Common investors can transmit shocks (investor similarity index constructed for testing). - China-related shocks can affect global risk premiums and trigger risk-off episodes. - Risk-off episodes: 15 distinct episodes since 1992 identified using a VIX-based rule; episodes 13 (21-Aug-2015) and 14 (15-Jan-2016) potentially involve China. - China holds reserves US$3.4 trillion as of end-2015 and is a major investor in U.S. Treasury securities; studies suggest China’s direct impact on U.S. long-term yields is likely limited. ### VII. Event study — Episodes and patterns - Recent high-impact episodes: - August 11, 2015: China announced a change to its exchange rate regime. - August 24, 2015: Chinese stock market fell by more than 8 percent (Black Monday). - January 4, 2016: renewed Chinese market volatility. - Methodology: - Identified 31 episodes where Shanghai Composite Index moved by more than 5 percent, excluding days with U.S. market movement > one standard deviation before Chinese open. - Identified 14 renminbi episodes since July 2005 where onshore renminbi-dollar rate moved by more than 0.5 percent in a day, excluding days with DXY movement > one standard deviation. - Event study results: - Average impact of China-related shocks on regional stock markets rose after the global financial crisis and further after June 2015. - Markets with strong trade links to China were more affected, particularly after the global financial crisis and after June 2015. - Regional currencies moved along with the renminbi, especially after August 2015. - Countries with strong trade links moved more closely with the renminbi. - China’s average impact on Asian equity markets was higher when global risk aversion was high (VIX above long-term average of 20); similar results for exchange rates. - Limitations: - Event selection cannot fully rule out regional or sectoral shocks causing co-movements. - Use of high-frequency (daily) data and identification of local news-triggered shocks mitigate contamination. - Formal empirical analysis corroborates event-study findings. ### VIII. Formal empirical approach and extensions - Two-step estimation (Forbes and Chinn (2004) framework extended): - Step 1: Estimate sensitivity (beta) of stock market returns to global, cross-country, and domestic factors using daily data. - Step 2: Treat estimated betas to center economies as dependent variables; regress on variables reflecting trade and financial linkages with center economies. - Extensions and enhancements: - Include China as a center country alongside Japan, euro area, and the United States. - Use value-added trade data (OECD TiVA) rather than gross trade to better capture final-demand exposures. - Use comprehensive financial linkage measures including portfolio, bank, and FDI claims; account for Hong Kong SAR as a financial gateway to China. - Sample period for daily empirical analysis: 2001 to 2014 (with event studies covering later turbulence). ### IX. Policy implications (summary) - For China: - Continued efforts to communicate policy intentions clearly and effectively are essential given rising influence on regional markets. - For other countries in the region: - Macroeconomic and macroprudential policies can build resilience against shocks emanating from China. - Over the long term, diversifying sources of growth in the region, including promoting services-sector growth, can reduce reliance on exports and diminish spillovers. *Source: Annex Table 3.2. Asian Bond Markets: Second-Stage Results––Determinants of Beta (content excerpt).* ### Annex 2 provides further details on the construction of the dataset used in the analysis. ### _wp16173 - Annex 2 provides further details on the construction of the dataset used in the analysis. ### Methodology (two-stage ICAPM-based approach) - First-stage regression (ICAPM): Rit = αi + Σc=1^4 βic Rct + Σg=1^3 γig Xgt + δi Yit + εi,t - Rit: return on country i's stock market index in day t. - Rct: return on center country c’ stock market index in day t. - Xgt: change in global factors in day t (global risk appetite, world interest rates, commodity prices). - Yit: changes in country risk in day t. - εi,t: normally distributed error term. - Returns are calculated as rolling-average, two-day returns. Returns are calculated in U.S. dollars and local currency; discussion focuses on local-currency absolute returns. Excess returns (adjusted for the risk-free rate) were also tested. - Second-stage regression (determinants of βic): - βic = υi + θ1 Tic + θ2 TXic + θ3 Fic + θ4 GFC + ηi,c - Tic: direct trade linkages between country i and c. - TXic: export competition in third markets between i and c. - Fic: direct financial linkages between i and c. - GFC: dummy equal to one for the period from 2008 to 2009. - ηi,c: normally distributed error term. - Estimation horizons: - Full sample: 2001 to 2014. - Subsamples: pre-GFC (2001–07) and post-GFC (2010–14). - Rolling 12-month samples starting from 2001. ### First-stage results (average market sensitivities) - China’s spillover coefficient is positive and significant for all economies in the sample and has increased in the post-GFC period. - U.S. spillover coefficient remains high and increased somewhat post-GFC. - Japan’s spillover coefficient declined in the post-GFC years and was similar to China’s as of end-2014. - Chow tests: changes in spillover coefficients from pre-GFC to post-GFC are statistically significant for China and Japan at the 1 percent level and for the United States at the 10 percent level. - Selected estimated beta coefficients for the full sample (illustrative; factor loadings shown): - China: 0.050*** - United States: 0.265*** - Japan: 0.241*** - Euro area: 0.045*** - India: China 0.080***; United States 0.162***; Japan 0.213***; Euro area -0.055** - Indonesia: China 0.075***; United States 0.198***; Japan 0.241***; Euro area -0.047** - Korea: China 0.038***; United States 0.099***; Japan 0.464***; Euro area 0.026 - Malaysia: China 0.049***; United States 0.106***; Japan 0.133***; Euro area 0.033** - New Zealand: China 0.022***; United States 0.213***; Japan 0.068***; Euro area 0.029*** - Philippines: China 0.040***; United States 0.305***; Japan 0.129***; Euro area 0.067*** - Taiwan Province of China: China 0.060***; United States 0.181***; Japan 0.352***; Euro area -0.002 - Thailand: China 0.073***; United States 0.116***; Japan 0.200***; Euro area -0.014 - Note: "*** p<0.01, ** p<0.05, * p<0.1"; Source: IMF staff estimates. ### Second-stage results (determinants of spillovers) - Main finding: trade exposure (Tic) is the primary transmission channel for spillovers from China into Asian equity markets. - Table 3 (selected coefficients, robust t-statistics in parentheses) - Linkages with China: - Trade exposure: 1.746* (2.058) - Trade competition: -0.128 (-0.251) - Financial linkages: 0.117 (0.542) - Observations: 126; R-squared: 0.241 - Linkages with Japan: - Trade exposure: 1.492 (0.799) - Trade competition: -1.057*** (-3.590) - Financial linkages: 0.271 (0.313) - Observations: 126; R-squared: 0.588 - Interpretation: - For China, trade exposure is statistically and economically significant; robustness checks confirm significance at the 5 and 1 percent levels. - For Japan, trade competition (export similarity) is the statistically significant channel—consistent with Japan’s role as an intermediate supplier versus China’s role as assembly hub. ### Changing role of financial linkages and Hong Kong SAR - Financial linkages with China became a significant transmission channel after the GFC. - Table 4 (interaction with Post-GFC dummy; robust t-statistics in parentheses): - Trade Linkages: 1.472* (2.080) - Trade Competition: -0.331 (-0.610) - Financial Linkages: Pre-GFC -0.003 (-1.179) - Financial Linkages: Post-GFC 0.463* (2.234) - Observations: 126; R-squared: 0.229; Number of id: 9 - Financial linkages with China are significant notably when measured including claims on Hong Kong SAR, underscoring Hong Kong SAR’s role as a financial gateway to China. - Robustness check excluding financial linkages through Hong Kong SAR: results remain broadly the same (trade exposure significant; financial linkages statistically insignificant in baseline). ### Risk-off scenarios and potential amplification - If a China-related shock is coupled with a “risk-off” episode (defined as a rise in the VIX by 10 percentage points), the impact on regional markets could be twice as much based on estimated sensitivities of those markets to other center economies. - Estimates obtained using βic coefficients from first-stage regressions and historical responses of center economies to a VIX shock of 10 percentage points. - China has two recent instances outside the sample where it may have contributed to a sharp rise in global risk aversion: August 2015 and January 2016. ### Sensitivity tests (robustness checks; summary of Table 5) - Categories of sensitivity tests: time fixed effects; different return definitions (dollar returns, excess returns); residual global factors; excluding Hong Kong SAR financial gateway; excluding Asian commodity exporters; controlling for country fundamentals-based risks; controlling for common investor channel. - Key outcomes (selected coefficients and significance; robust t-statistics in parentheses): - Trade exposure coefficients across specifications: - Controlling for time fixed effects: 2.275** (2.918) - Dependent variable: dollar returns: 2.428** (2.474) - Dependent variable: excess returns: 2.428** (2.473) - Independent variable: residual global factors: 2.005** (2.665) - Excluding financial linkages through Hong Kong SAR: 1.898* (2.097) - Excluding Asian commodity exporters: 3.015*** (6.039) - Controlling for country fundamentals-related risks: 1.970** (2.371) - Controlling for common investor channel: 1.952** (2.405) - Trade competition: - Varied signs; statistically significant negative in some specifications (e.g., -1.524* (-2.059) with dollar returns). - Financial linkages: - Often statistically insignificant at the 10 percent level in baseline and many robustness checks; some larger but insignificant estimates in alternate specifications. - Country fundamentals-based risk index: - Coefficient: 0.028* (2.239) when included. - Investor similarity index: - Coefficient: 0.013 (0.080); not statistically significant. - Excluding net commodity exporters (Australia, Indonesia, Malaysia, New Zealand) increased the trade linkages coefficient materially (statistically significant at the 1 percent level and almost twice the baseline). The coefficient on trade competition became positive and significant at the 10 percent level in that subsample. - Use of residual global factors (VIX, world interest rates, commodity prices orthogonalized to center returns) does not materially change results; correlation structure does not point to significant multicollinearity. ### Conclusions and policy implications - Empirical summary: - Financial spillovers from China to regional markets were already increasing before the recent turbulence. - Main transmission channel: trade linkages; direct financial linkages increased in importance post-crisis. - Without an impact on global risk premiums, spillovers from China are not yet at U.S. levels but are comparable to Japan. - Short-term exposures: - Economies most likely affected: those with strong trade links to China; Singapore, Korea, and Taiwan Province of China due to strong financial links. - China-related shocks that induce global “risk-off” episodes can affect Japan (via safe-haven flows), and Indonesia and Malaysia (currency sensitivity). - Long-term outlook: - A more balanced Chinese economy can benefit the region and world over the long term. - Capital market liberalization in China could improve asset diversification, social safety nets, private consumption, and imports. - Opportunities from China’s rebalancing include benefits to lower-income countries as China moves up the value chain and increases outward FDI; services import demand (e.g., tourism) could partly offset negative trade-channel impacts. - Policy recommendations: - For China: continue clear and effective communication of policy intentions given rising regional influence. - For other countries: use policy buffers judiciously to discharge macroeconomic support measures where available. - Employ macroprudential policies to safeguard financial stability if volatile asset prices lead to substantial capital outflows or worsen vulnerabilities (for example in corporate sectors). - Pursue structural reforms to diversify growth sources, including promoting the services sector to reduce reliance on exports. *Source: IMF staff estimates.* ### ANNEX 1. EVENT STUDY: EXCEPTIONALLY LARGE MOVEMENTS IN CHINESE MARKETS ### ANNEX 1. EVENT STUDY: EXCEPTIONALLY LARGE MOVEMENTS IN CHINESE MARKETS ### Identification of shocks originating in China (stock market) - Exceptionally large movement defined as a daily change in the Shanghai Composite Index by more than 5 percentage points. - Exclusions to isolate domestic shocks: - Days during the global financial crisis (2008–09). - Days when the U.S. stock market moves by more than one standard deviation just hours before the Chinese market opens (used as a proxy for global events). - For remaining days, domestic news items were searched to identify drivers of large movements. - Based on this selection criterion, 31 episodes were identified from January 2001 to June 2016 (Annex Table 1.1). - Selected notable episodes (date — Chinese stock return — event summary): - 7/30/2001 — -5.3 — Regulators issue rules ordering listed firms to sell state shares in IPOs; sparks four-year slump. - 1/23/2002 — 6.3 — Market rises on news selling of state shares may be delayed. - 2/27/2007 — -8.8 — Government imposes controls to curb speculation; triggers ~9 percent domestic drop and worldwide losses ~2 percent. - 5/30/2007 — -6.5 — Ministry of Finance raises trading tax to 0.3 percent from 0.1 percent; index falls 21 percent by June 5. - 1/19/2015 — -7.7 — Fall following tighter rules for margin lending; reports of PBoC injecting liquidity via MLF rollovers. - 6/26/2015 — -7.4 — Morgan Stanley warns shares overvalued; forecasts possible 30% fall through mid-2016. - 7/27/2015 — -8.5 — Biggest daily drop since 2007 after industrial profits fall and local government debt estimate RMB 30 trillion (US$ 4.9 trillion) at end-2014. - 8/24/2015 — -8.5 — “Black Monday”: equities tumble 8.5 percent; U.S. equity volatility surges; VIX quotations suspended in early session. - 1/4/2016 — -6.9 — Chinese stocks fall sharply by 7 percent triggering circuit breakers. - 2/25/2016 — -6.4 — Biggest loss in one month amid increased funding costs and reports fiscal deficit could widen. - Sources for events: Bloomberg L.P.; and news reports. - Note abbreviations preserved: CFSC, CSRC, GFC, IPO, MLF, PBoC, PMI. ### Identification of shocks originating in China (onshore renminbi exchange rate) - Exceptionally large movement defined as a daily change in the onshore renminbi-dollar exchange rate by more than 0.5 percentage points. - Exclusions to isolate domestic currency shocks: - Days when the DXY dollar index moves by more than one standard deviation against major currencies (to exclude movements driven by the U.S. dollar rather than the renminbi). - DXY is defined as an index of the value of the U.S. dollar relative to a basket of six currencies: the euro, the Japanese yen, the pound sterling, the Canadian dollar, the Swedish krona, and the Swiss franc. - For remaining days, domestic news items were searched to identify drivers of large exchange rate movements. - Based on this selection criterion, 14 episodes were identified from January 2001 to June 2016 (Annex Table 1.2). - Selected notable episodes (date — change in the CNY/USD rate (percent) — event summary): - 7/21/2005 — 2.1 — China announces adoption of a managed floating exchange rate regime, with revaluation of renminbi by 2.1 percent. - 4/7/2015 — -0.9 — Renminbi depreciates after PBoC cuts reverse repo rate 10 bps to 3.45%; junk bond yields jump after Cloud Live Technology Group defaults RMB 241 mn ($39 mn). - 8/11/2015 — -1.8 — China introduces a new exchange rate mechanism; PBoC lowers renminbi fixing rate by 1.9 percent. - 8/12/2015 — -1.0 — Currency falls to four-year low, second day of decline after new mechanism. - 1/7/2016 — -0.6 — PBoC fixes onshore yuan 0.5% weaker; FX reserves record largest ever monthly decline in December, falling US$108 billion. - 2/15/2016 — 1.2 — Renminbi rallies the most since 2005 after PBoC Governor voices support as markets reopen after Lunar New Year. - Sources for events: Bloomberg L.P.; and news reports. ### ANNEX 2. CONSTRUCTION OF THE DATASET USED IN THE EMPIRICAL ANALYSIS ### Coverage and data frequency - Economies covered: - Nine Asian economies: Australia, India, Indonesia, Korea, Malaysia, New Zealand, the Philippines, Taiwan Province of China, and Thailand. - Four “center” economies: China, Japan, the euro area, and the United States. - Daily data for the first-stage regression compiled from Bloomberg. - Annual data for the second-stage regression assembled from multiple sources detailed below. ### First-stage variables and measurement - Asset returns (R_it): stock returns measured in either local currency or U.S. dollars based on stock indices compiled by Bloomberg. - Stock indices used: ASX 200 (Australia), Shanghai Composite Index (China), Eurofirst 300 (the euro area), BSE SENSEX 30 (India), Jakarta Composite (Indonesia), NIKKEI 225 (Japan), KOSPI (Korea), FTSE/KLCI (Malaysia), NZX 50 (New Zealand), PSE Composite Index (the Philippines), TWSE (Taiwan Province of China), SET Index (Thailand), and S&P500 (the United States). - Stock market returns are calculated as rolling-average, two-day returns (average two-day returns) to adjust for non-overlapping market hours. - Global risk appetite: CBOE Volatility Index (VIX index). - World interest rates: “shadow” U.S. policy rate estimated by Wu and Xia (2016). - Commodity prices: Bloomberg Commodity Index (covers 22 commodities in seven sectors). - Country risk: country-specific credit default swap (CDS) spreads or, if unavailable, J.P. Morgan Emerging Market Bond Index-Global (EMBIG) sovereign spread. ### Second-stage variables and construction (annual) - Direct trade linkages (T_ic): - Data source: OECD-WTO Trade in Value Added (TiVA) database for years 2000, 2005, and 2008–11; continuous series through 2014 constructed using UN Comtrade and national income accounts, following OECD-WTO methodology. - Definition preserved exactly: - T_i,c = E_i^c / GDP_i - where E_i^c is the value added produced by country i and exported for final demand in country c (includes (i) direct exports of final goods and services from i to c; and (ii) exports of intermediate goods from i to other countries that will eventually be re-exported to c for final demand). - Trade competition (export similarity index, TX_i,c): - Formula preserved exactly: - TX_i,c = sum_{k=0}^n [ min(EX_i,k, EX_c,k) ] - where EX_i,k and EX_c,k are industry k’s export shares in country i’s and country c’s exports. - Product-level data: UN Comtrade five-digit product level. - Direct financial linkages (F_i,c): - Formula preserved exactly: - F_i,c = (P_i,c + B_i,c + D_i,c) / GDP_i - where P_i,c = portfolio claims, B_i,c = cross-border bank claims, D_i,c = direct investments of country i in country c. - Data sources: - Portfolio claims: IMF Coordinated Portfolio Investment Survey (CPIS). - Cross-border bank claims: unpublished locational banking statistics from the Bank for International Settlements (BIS). - Direct investments post-2009: IMF Coordinated Direct Investment Survey (CDIS); prior to 2009: UNCTAD. - Financial linkages with China include claims on Hong Kong SAR to account for its role as a financial gateway to China; robustness checks performed with and without Hong Kong SAR inclusion. - Empirical note: about 50 to 60 percent of the Hang Seng Index is comprised of mainland companies listed in Hong Kong SAR (text chart data cited). - GDP used as denominator: IMF World Economic Outlook (reported in U.S. dollars). ### Summary statistics and correlations (as reported) - Annex Table 2.1: Correlation between the variables in first stage (selected entries preserved exactly): - SMI CHN — SMI USA: 0.17 - SMI CHN — SMI JPN: 0.24 - SMI CHN — SMI EA: 0.14 - VIX correlations: with SMI CHN = -0.07; with SMI USA = 0.10; with SMI JPN = -0.14; with SMI EA = 0.03 - Commodity Prices correlations: with SMI CHN = 0.11; with VIX = -0.24 - Country Risk correlations: with SMI CHN = -0.09; with VIX = 0.23 - Shadow Fed Funds Rate correlations: with SMI CHN = 0.01; with Country Risk = -0.06 - Annex Table 2.2: Correlation between the variables in second stage (preserved exactly): - Trade Linkages — Trade Competition: 0.16 - Trade Linkages — Financial Linkages: 0.29 - Trade Competition — Financial Linkages: 0.01 ### ANNEX 3. EMPIRICAL RESULTS: BOND MARKETS ### First-stage regression results (daily changes in 10-year government bond yields, percentage points) - General finding: Asian bond markets’ sensitivities to the Chinese bond market remain low except for a few exceptions (Korea and Taiwan Province of China). Regional bond markets are more sensitive to the U.S. bond market and, to a lesser extent, to the Japanese bond market. - Factor loadings for the full sample (only factor loadings shown for illustrative purposes; columns: China / United States / Japan; N and R2 shown where present): - Australia: 0.003 / 0.522*** / 0.531*** — N = 2,421 — R2 = 0.391 - India: 0.053 / 0.089*** / 0.087 — N = 2,421 — R2 = 0.016 - Indonesia: 0.137 / -0.041 / -0.087 — N = 2,421 — R2 = 0.009 - Korea: 0.077** / 0.192*** / 0.153*** — N = 2,421 — R2 = 0.090 - Malaysia: 0.023 / 0.079*** / 0.035 — N = 2,421 — R2 = 0.019 - New Zealand: -0.005 / 0.347*** / 0.269*** — N = 2,113 — R2 = 0.283 - Philippines: 0.066 / -0.129*** / 0.000 — N = 2,207 — R2 = 0.010 - Taiwan Province of China: 0.045*** / 0.085*** / 0.133*** — N = 2,421 — R2 = 0.115 - Thailand: 0.089** / 0.146*** / 0.093* — N = 2,421 — R2 = 0.041 - Note on significance: *** p<0.01, ** p<0.05, * p<0.1 - Source: IMF staff estimates. ### Second-stage regression results (determinants of Asian bond market sensitivities) - Main finding: In contrast to equity market results, all three explanatory variables (trade linkages, trade competition, financial linkages) lose their degree of significance for bond markets. Even with alternate specifications, results for Chinese bond market specifications remain insignificant. - Reported regression snippet (Systemic Economy/Region (centers) — coefficients and statistics preserved exactly): - Trade exposure: 0.004 0.128** (robust t-statistics in parentheses: (0.167) (2.521)) - Trade competition: -0.021 -0.000 (robust t-statistics: (-1.186) (-0.017)) - Financial linkages: 0.006 0.040 (robust t-statistics: (0.416) (0.827)) - Observations: 9090 - R-squared: 0.119 0.193 - Note: Robust t-statistics reported in parentheses. Significance notation preserved: *** p<0.01, ** p<0.05, * p<0.1. - Source: IMF staff estimates. *Source: ANNEX 1–3 as presented in the provided content unit.* ### Annex Table 3.2. Asian Bond Markets: Second-Stage Results—Determinants of Beta Coefficients ### Annex Table 3.2. Asian Bond Markets: Second-Stage Results—Determinants of Beta Coefficients ### References - Ahuja, A., and M. Nabar. 2012. “Investment-Led Growth in China: Global Spillovers.” IMF Working Paper 12/267, International Monetary Fund, Washington. - Aizenman, J., M. D. Chinn, and H. Ito. 2015. “Monetary Policy Spillovers and the Trilemma in the New Normal: Periphery Country Sensitivity to Core Country Conditions.” NBER Working Paper No. 21128. National Bureau of Economic Research, Cambridge, Massachusetts. - Bayoumi, T. and F. Ohnsorge. 2013. “Do Inflows or Outflows Dominate? Global Implications of Capital Account Liberalization in China.” IMF Working Paper 13/189, International Monetary Fund, Washington. - Brooks, R. and M. Del Negro. 2003, “Firm-level Evidence on Globalization,” IMF Working Paper 03/55, International Monetary Fund, Washington. - Botman, D., I. de Carvalho Filho, R. Lam. 2013. “The Curious Case of the Yen as a Safe Haven Currency: A Forensic Analysis.” IMF Working Paper 13/228, International Monetary Fund, Washington. - Cashin, P., K. Mohaddes, and M. Raissi. 2016. “China’s Slowdown and Global Financial Market Volatility: Is World Growth Losing Out?” IMF Working Paper 16/63, International Monetary Fund, Washington. - Claessens, S., H. Tong, and I. Zuccardi. 2012. “Did the Euro Crisis Affect Non-Financial Firm Stock Prices through a Financial or Trade Channel?” IMF Working Paper No. 11/227, International Monetary Fund, Washington. - De Bock, R., and I. de Carvalho Filho. 2015. “The Behavior of Currencies During Risk-off Episodes.” Journal of International Money and Finance 53: 218–34. - Diebold, F. X., and K. Yilmaz. 2012. “Better to Give than to Receive: Predictive Directional Measurement of Volatility Spillovers.” International Journal of Forecasting 28(1): 57–66. - Dizioli, A., J. Guajardo, V. Klyuev, R. Mano, and M. Raissi. 2016, “Spillovers from China’s Growth Slowdown and Rebalancing to the ASEAN-5 Economies” IMF Working Paper, International Monetary Fund, Washington. - Duval, R., K. Cheng, K. H. Oh, R. Saraf, and D. Seneviratne. 2014. “Trade Integration and Business Cycle Synchronization: A Reappraisal with Focus on Asia.” IMF Working Paper 14/52, International Monetary Fund, Washington. - Duval, R., N. Li, R. Saraf, and D. Seneviratne. 2015. “Value-Added Trade and Business Cycle Synchronization.” Journal of International Economics 99: 251–62. - Elekdag, S., P. Rungcharoenkitkul, and Y. Wu. 2012. “The Evolution of Asian Financial Linkages: Key Determinants and the Role of Policy.” IMF Working Paper 12/262, International Monetary Fund, Washington. - Frankel, Jeffrey. 1994. “Introduction,” in The Internationalization of Equity Markets, edited by Jeffrey Frankel, University of Chicago Press, Chicago and London. - Forbes, Kristin. 2013. “The Big-C: Identifying and Mitigating Contagion,” Jackson Hole Conference Proceedings. - Forbes, K.J., and M. Chinn. 2004. “A Decomposition of Global Linkages in Financial Markets Over Time.” Review of Economics and Statistics 86(3): 705–22. - Forbes, K. J., and R. Rigobon. 2002. “No Contagion, Only Interdependence: Measuring Stock Market Comovements.” Journal of Finance, 2002, v57, 2223–61. - Goldman Sachs. 2016. “Walled In: China’s Great Dilemna.” Investment Strategy Group, January. - Gong, C. and S. Kim. 2013. “Economic Integration and Business Cycle Synchronization in Asia.” Asian Economic Papers, Vol. 12 No. 1, pp. 76–99. - Guimarães-Filho, R., and G. Hong. 2016. “Dynamic Connectedness of Asian Equity Markets.” IMF Working Paper 16/57, International Monetary Fund, Washington. - Hong, G., J. Lee, W. Liao, and D. Seneviratne. 2016. “China and Asia in Global Trade Slowdown.” IMF Working Paper 16/105, International Monetary Fund, Washington. - Institute of International Finance (IIF). 2016. “Capital Flows to Emerging Markets.” - International Monetary Fund. 2013. “Dancing Together? Spillovers, Common Shocks, and the Role of Financial and Trade Linkages” in World Economic Outlook, Washington, October. - International Monetary Fund. 2014. Global Liquidity, Policy Paper, Washington, April. - International Monetary Fund. 2015. “People’s Republic of China—2015 Article IV Consultation—Staff Report.” IMF Country Report 15/234, Washington. - International Monetary Fund. 2016a. World Economic Outlook Update. Washington, January. - International Monetary Fund. 2016b. “The Growing Importance of Financial Spillovers from Emerging Market Economies” in Global Financial Stability Report. Washington, April. - International Monetary Fund. 2016c. Regional Economic Outlook: Asia and the Pacific. Washington, April. - Kireyev, A., and A. Leonidov. 2016. “China’s Import Slowdown: Spillovers, Spillins, and Spillbacks.” IMF Working Paper 16/51, International Monetary Fund, Washington. - Kho, B-C, D. Lee, and R. M. Stulz. 2000. “U.S. Banks, Crises, and Bailouts: From Mexico to LTCM,” American Economic Review 90, 28–31. - Kubelec, C. and F. Sá. 2008. “Financial Globalization and Cross-Country Spillovers.” mimeo, Bank of England. - Prachi M., K. Moriyama, P. N. Diaye, and L. Nguyen. 2014. “Impact of Fed Tapering Announcements on Emerging Markets.” IMF Working Paper 14/109, International Monetary Fund, Washington. - Rhee, Changyong. 2015. “Managing China’s Economic Transition.” iMFdirect, October 5. Available at https://blog-imfdirect.imf.org/2015/10/05/managing-chinas-economic-transition/ - Sahay, R., V. Arora, T. Arvanitis, H. Faruqee, P. N. Diaye, T. Mancini-Griffoli, and an IMF Team, 2014. “Emerging Market Volatility: Lessons from the Taper Tantrum.” IMF Staff Discussion Note SDN 14/09, September 2014. - Shaghil A., B. Coulibaly, and A. Zlate. 2015. “International Financial Spillovers to Emerging Market Economies: How Important Are Economic Fundamentals?” International Finance Discussion Papers 1135. Board of Governors of the Federal Reserve System, Washington. - Stulz, René M. 1999. “International Portfolio Flows and Security Markets,” in International Capital Flows, edited by Martin Feldstein, University of Chicago Press, Chicago. - Warnock, F. E. and V. C. Warnock. 2005. “International Capital Flows and U.S. Interest Rates.” Board of Governors of the Federal Reserve System. International Finance Discussion Papers Number 840. September 2005. - Wu, J. C. and Xia, F. D. 2016. “Measuring the Macroeconomic Impact of Monetary Policy at the Zero Lower Bound” Journal of Money, Credit, and Banking, 48(2-3), 253-291. *Annex Table 3.2. Asian Bond Markets: Second-Stage Results—Determinants of Beta Coefficients (source document contents).* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16173.pdf_