## _wp09166

## Source details

**Canonical URL:** [_wp09166](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09166.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09166.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09166.pdf.json)

---

### I. Introduction and Research Objective
- Research question:
  - Whether the U.S. subprime financial turmoil had any statistically significant effect on both the daily returns and the conditional volatility of stock prices in mainland China (China) and Hong Kong SAR (HK).
- Methodology overview:
  - Two-stage procedure:
    - Stage 1: Estimate a group of univariate GARCH models (UGARCH) — AR(1)-GARCH(1,1) with a subprime event indicator.
    - Stage 2: Estimate a group of multivariate GARCH models (MGARCH) — Diagonal VECH (identical to Diagonal BEKK) to identify sources and magnitudes of price and volatility spillovers.

### Key empirical findings (summary)
- Using both UGARCH and MGARCH models:
  - China’s stock market is not immune; price and volatility spillovers from the United States to China are evidenced in MGARCH models.
  - HK’s equity returns show more significant price and volatility spillovers from the United States than China; past U.S. volatility shocks have a more persistent effect on future volatility in HK than in China.
  - Cross-volatility between HK and China is lower than between the United States and China, but the conditional correlation between China and HK outweighs their correlations with the United States, indicating increasing financial integration between China and HK.
  - Subprime turmoil increased conditional volatility in China’s financial sector and HK-listed companies; HK experienced both higher volatility and lower returns associated with subprime events.

### Stylized facts and market performance (selected exact figures)
- By end-October 2008 stock prices in major economies had declined at least 30 percent since their peak in October 2007.
- Specific declines:
  - China experienced a decline of 71 percent.
  - HK experienced a decline of 56 percent.
- Capital flow and hot money movements:
  - China experienced a hot money inflow of about USD 130 billion in the first two quarters of 2008.
  - Documented outflow values include an outflow of about USD 7.2 billion and 90 billion (Figure 3) — hot money defined as the residual of foreign exchange reserves minus trade surplus and foreign direct investment.
  - Individual reported bank exposures (China’s Securities Daily, Sep 2008): $191.4 million at China Construction Bank (CCB), $151.8 million at the Industrial and Commercial Bank of China (ICBC), $128.8 million at the Bank of China (BOC), $76 million at the CITIC Bank, $70 million at the China Merchants Bank, $70.02 million at the Bank of Communications and $33.6 million at the Industrial Bank.
  - Direct exposure of the Chinese financial system to U.S. structured credit products: about $10 billion.
  - Direct exposure to subprime-related assets accounted for 3.7 percent of total banking assets (CBRC, 2008).
  - Seven Chinese listed banks announced total bond holdings of $721 million in Lehman Brothers as of September 22, 2008.
  - HK banks’ total exposure to U.S. subprime securities and structured assets remained well below ½ percent of banking system assets (IMF, 2008a).

### Transmission channels from the United States to China and HK
- Loss exposures (direct):
  - China: about $10 billion direct exposure to U.S. structured credit products; subprime-related assets = 3.7 percent of total banking assets.
  - HK: exposures well below ½ percent of banking system assets.
  - Seven Chinese listed banks held $721 million in Lehman Brothers bonds (as of September 22, 2008).
- Loss of confidence (indirect):
  - Indirect influence through confidence shocks as negative news accumulated (Lehman bankruptcy, Merrill Lynch sale, AIG rescue).
- Capital flow ambiguity:
  - Two opposing effects:
    - Credit crunch in advanced economies could drive foreign capital out of China and HK to meet liquidity needs.
    - Some foreign investors might view China and HK as relatively resilient and continue or increase investment.

### Data and econometric framework (exact specifications)
- Data:
  - Daily close-of-day stock market price indices based on national currencies for the United States, China, and HK.
  - Returns: R_t = LnP_t - LnP_{t-1}. (Price returns are not corrected for dividends.)
  - Sample period: January 1, 2005 to October 31, 2008.
  - Subprime turmoil emphasis sample: January 1, 2007–October 31, 2008.
- Domestic control variables:
  - Interest rate differential, surprises in M2 growth, CPI inflation, industrial production; surprises in monthly changes of China’s trade balance; difference between HK’s export and import growth as proxy for HK trade balance.
  - Surprise measures: difference between official data on release dates (real-time) and market forecasts; surprise value used on release date, 0 for other days.
- Global variables:
  - Chicago Board Options Exchange’s Volatility Index (VIX) and Lehman Brothers Swaption Volatility Index (LBSPX — weighted average of 1-6 month swaption volatilities).
- UGARCH specification (examples preserved):
  - Return equation example: R_t = constant + ζ_t R_{t-1} + λX_{t-1} + φV_t^f + ð event_t + Є_t.
  - Variance equation: σ_t^2 = α_0 + α_1 Є_{t-1}^2 + β_1 σ_{t-1}^2 + μ event.
- MGARCH specification:
  - Diagonal VECH (A and B restricted to diagonals). Conditional mean includes own lagged return and foreign lagged returns: R_t = constant + ζ_t R_{t-1} + η R_{f,t-1} + λ X_{t-1} + φ V_t^f + Є_t.
  - Variance: σ_t^2 = α_0 + α_1 Є_{t-1}^2 + β_1 σ_{t-1}^2.
  - Errors assumed Student's t-distribution due to excess kurtosis.
- Estimation:
  - GARCH(1,1) and Diagonal VECH estimated by maximum likelihood using BHHH algorithm.

### Diagnostics and preliminary data properties
- ADF tests:
  - Log change in stock price series I(0).
  - VIX and LBSPX I(1).
- Model diagnostics:
  - Squared returns show persistence and clustering; ARCH tests support GARCH formulation.
  - Distribution of squared returns skewed and leptokurtic; standardized residuals show excess kurtosis.
  - Information criteria (Akaike and Schwarz) suggest one lag of dependent variable and regressors.

### UGARCH empirical findings — China (exact findings and coefficients where stated)
- Dependent variables: Shanghai Composite Index, Shanghai financial index, FXI (iShare FTSE/Xinhua China 25 index fund; priced in U.S. dollars; includes 11 large financial corporations).
- Mean equation findings:
  - FXI intercept significant at the 10 percent level with negative sign, indicating FXI declines at a pace of 0.02 percent per day on average, holding other things equal.
  - Lagged daily return significant in FXI model.
  - Interest rate differential significant and negative for Shanghai Composite Index and FXI.
  - Most other domestic macro surprises statistically insignificant.
  - Global market volatility variables and subprime events generally statistically insignificant in mean equations for all three China models.
- Variance equation findings:
  - Coefficients of ARCH, GARCH and subprime events are positive and statistically significant in all three China models.
  - α_1 + β_1 was less than 1 in all cases, producing finite unconditional variances.
  - Subprime events variables significant in Shanghai Composite Index and FXI variance models — volatility of the financial sector and FXI indices responded to external subprime shocks.

### UGARCH empirical findings — Hong Kong (exact findings)
- Dependent variables: Hang Seng index and Hang Seng financial index.
- Mean equation findings:
  - Intercept term for all HK models is positive, reflecting HK’s stock prices increase at a pace of 0.01 percent per day on average, holding other things equal.
  - Interest rate differentials statistically significant with positive signs.
  - VIX significant in both HK models with negative signs.
  - Subprime event variable significant in the mean equation with negative sign — subprime events materially reduced levels of HK stock price returns.
- Variance equation findings:
  - Subprime event variable statistically significant with positive sign — subprime events increased conditional volatility of HK stock price returns.
- Overall: subprime events had material negative effects on both conditional volatility and levels of HK stock price returns.

### MGARCH findings, volatility and spillovers (preserved numeric values)
- Own-volatility (ARCH) spillovers (composite indices, Ai values):
  - United States: 0.04
  - China: 0.13
  - HK: 0.16
- Cross-volatility effects (composite indices):
  - U.S. ↔ China: 0.07
  - U.S. ↔ HK: 0.08
  - HK ↔ China: 0.15
- Lagged volatility persistence (GARCH term β) for composite indices (Bi values):
  - United States: 0.97 (Jan. 1, 2007-Oct. 31, 2008 table: Bi1 = 0.9707***/0.9707***)
  - HK: 0.83
  - China: insignificant (text)
- Cross-volatility persistence:
  - U.S. → China: 0.40
  - U.S. → HK: 0.90
  - HK ↔ China: 0.37
- Selected Diagonal VECH A and B elements (Table 9 examples, values preserved):
  - Ai1 examples: 0.0408***(0.0094), 0.0733***(0.0162), 0.0808***(0.0144)
  - Ai2 examples: 0.0733***(0.0162), 0.1315***(0.0496), 0.145***(0.036)
  - Ai3 examples: 0.0808***(0.0144), 0.145***(0.036), 0.1598***(0.038)
  - Bi1 examples: 0.9707***(0.0073), 0.4028***(0.1424), 0.8987***(0.0182)
  - Bi2 examples: 0.4028***(0.1424), 0.17(0.1182), 0.3729***(0.1364)
  - Bi3 examples: 0.8987***(0.0182), 0.3729***(0.1364), 0.832***(0.0329)
- Table 10 (Variance Covariance equations — Jan. 1, 2007-Oct. 31, 2008) selected α and αi coefficients:
  - α0: 0.0002 (0.0005)
  - αi1: -0.0525 (0.0428), 0.2461***(0.0423), 0.3353***(0.0259)
  - αi2: 0.0225(0.014), 0.0000(0), 0***(0)
  - αi3: -0.1591***(0.04), -0.0075(0.0525), -0.0510(0.0389)
- MGARCH mean spillover indication:
  - Price spillovers from the United States to HK more significant than to China with a coefficient of 0.27 to China and 0.76 to HK (text fragments indicate 0.27 to China and 0.76 to HK; exact phrasing in source: "price spillovers from the United States to HK have been more significant than to China with a coefficient of 0.27 to China and 0.76 to HK. This may reflect China’s less financial openness than HK.").

### Conditional correlations (Jan. 1, 2007-Oct. 31, 2008)
- Relative magnitudes (qualitative and directional, preserved wording):
  - Cor(LOGSPX,LOGSHCOMP) — correlation between U.S. S&P 500 index and Shanghai Composite index: much lower than with HK.
  - Cor(LOGSPX,LOGHSI) — correlation between U.S. S&P 500 index and Hang Seng index: higher than with Shanghai.
  - Cor(LOGSHCOMP,LOGHSI) — correlation between Shanghai Composite index and Hang Seng index: highest among the three markets.
- Interpretation:
  - Conditional correlation analysis indicates limited openness in China and increasing financial integration between China and HK.

### Robustness checks (preserved outcomes)
- Variations tested:
  - UGARCH and MGARCH models using financial sector stock prices.
  - Extended starting sample date from January 1, 2007 to January 1, 2005.
  - Additional dependent and explanatory variables and various changes in estimation methodology.
- Robustness results:
  - Mean and variance equation results generally hold for alternative specifications.
  - Minor differences: mean spillover and ARCH coefficients slightly lower, GARCH effects higher in financial-sector specifications; China’s own-volatility coefficient becomes significant in some specifications.
  - Overall results are rather robust with respect to sample period, additional variables, and estimation methodology.

### Policy conclusions and implications (verbatim positions and recommendations preserved)
- Main empirical conclusions:
  - There is no “decoupling” story for China and HK. No economy can be totally immune to the subprime financial turmoil. Regression results strongly support this.
  - China and HK financial markets are very much influenced by U.S. monetary and financial conditions; increasing integration implies authorities should be alert to negative spillover effects.
  - Greater attention is needed to the risk of a virulent feedback loop between the financial markets and the economy.
  - International policy coordination is increasingly important because markets transmit volatility rapidly, leaving little time for policy intervention; recent policy actions by both advanced and emerging market central banks (including China) underline this need.
  - Mainland and HK authorities need further cooperation given increasing volatility spillovers between China and HK’s stock markets.
- Recommendations concerning capital account liberalization and domestic reforms:
  - Chinese authorities should be more cautious in their approach to capital account liberalization.
  - As de facto capital account openness increases, authorities should strengthen financial markets and relevant infrastructure (governance, the exchange rate system, the supervisory system, etc.) alongside the natural opening-up process.
  - A pragmatic approach to capital account liberalization is advocated given the era of financial globalization.

### Appendix A — Timeline highlights and key quantitative indicators (selected exact figures)
- Selected major event dates and figures (non-exhaustive):
  - 2006-12-28: Ownit Mortgage Solutions files for bankruptcy.
  - 2007-6-22: Bear Stearns pledges up to $3.2 billion to bail out one of its hedge funds.
  - 2007-7-18: Bear Stearns hedge funds lost over 90 percent of their value, equal to over $1.4 billion.
  - 2007-10-5: Merrill disclose $5.5bn in subprime loss.
  - 2007-10-9: US subprime losses may reach $150 billion (S&P).
  - 2007-11-12: Subprime losses may reach $400 billion (analysts expected).
  - 2008-1-15: Citigroup announced mortgage portfolio dropped in value by $18.1 Billion.
  - 2008-3-31: US Treasury backs regulatory overhaul, broader Fed role.
  - 2008-3-13: Subprime losses may reach 285 billion dollars (S&P).
  - 2008-4-8: IMF says writedowns on US assets could total $945bn.
  - 2008-4-15: OECD raises subprime loss tally to $350-420 billion.
  - 2008-6-18: Paulson & Co. says writedowns may reach $1.3 Trillion.
  - 2008-8-7: “Yes, That's $2 Trillion of Debt-Related Losses.”
  - 2008-10-6 week: Dow Jones lost 22.1 percent that week, down 40.3 percent since October 9, 2007; S&P 500 lost 18.2 percent that week, down 42.5 percent since October 9, 2007.
- Foreclosure and delinquency indicators:
  - RealtyTrac: foreclosure filings surged 90 percent in May (2007) from May 2006; foreclosure filings surged to 243,000 in August (2007), up 115 percent from August 2006.
  - Mortgage Bankers Association: delinquency rate 5.12 percent in Q2 2007.
  - By end-2007, 2.04 percent of all mortgages were in the foreclosure process.
- Aggregate loss estimates (selected):
  - 2007-10-9: $150 billion.
  - 2007-11-12: $400 billion.
  - 2007-11-20: $270 billion.
  - 2007-11-22: $300 billion (OECD estimate).
  - 2008-2-29: $600 billion (UBS analysts report).
  - 2008-4-8: $945bn (IMF).
  - 2008-5-14: US$400 billion (Fitch).
  - 2008-6-18: $1.3 Trillion (Paulson & Co.).
  - 2008-8-7: $2 Trillion (quoted phrase).
  - 2008-10-28: '£1.8 trillion' (Bank of England world credit loss estimate).

### Appendix B — FXI US Equity members (constituents listing preserved)
- Constituents (1–25) by sector and name (preserved order and labels):
  - 1 Non-financial China Mobile Ltd
  - 2 Financial Industrial & Commercial Bank of China - H
  - 3 Financial China Life Insurance Co-H
  - 4 Non-financial PetroChina Co Ltd-H
  - 5 Non-financial CNOOC Ltd
  - 6 Financial PING AN Insurance Group Co-H
  - 7 Financial Bank of China Ltd - H
  - 8 Non-financial China Petroleum & Chemical Co-H
  - 9 Non-financial China TeleCom CoRP Ltd-H
  - 10 Financial China Construction Bank-H
  - 11 Financial China Merchants Bank - H
  - 12 Financial Bank of Communication Co-H
  - 13 Financial Bank of China HONG KONG Holding Ltd
  - 14 Non-financial China UNICOM Ltd
  - 15 Non-financial China Shenhua Energy Co - H
  - 16 Non-financial China Coal Energy Co - H
  - 17 Non-financial China Communications Const-H
  - 18 Financial China Merchants Holdings International
  - 19 Non-financial China NETCOM Group Corp HK Ltd
  - 20 Non-financial China COSCO Holdings-H
  - 21 Non-financial Huaneng Power International Inc-H
  - 22 Non-financial Datang International Power Generating Co-H
  - 23 Non-financial Aluminum Corp of China Ltd-H
  - 24 Financial China CITIC Bank - H
  - 25 Non-financial AIR China Limited-H
- Data provenance: Bloomberg L.P.

*Source: Content drawn from the provided IMF working paper excerpt.*

### References..............................................................................................................

### _wp09166 - References

### I. Introduction and Research Objective
- The paper examines whether the U.S. subprime financial turmoil has had any statistically significant effect on both the daily returns and the conditional volatility of stock prices in mainland China (China) and Hong Kong SAR (HK).
- Methodology:
  - Two-stage procedure:
    - Stage 1: Estimate a group of univariate GARCH models (UGARCH).
    - Stage 2: Estimate a group of multivariate GARCH models (MGARCH) to identify sources and magnitudes of spillovers.

### Key empirical findings
- Using both UGARCH and MGARCH models, the paper finds:
  - (i) China’s stock market has not been immune to the financial crisis and there is no decoupling story, as clearly evidenced by the price and volatility spillovers from the United States to China in MGARCH models.
  - (ii) HK’s equity returns exhibit more significant price and volatility spillovers from the United States, and past volatility shocks in the United States have a more persistent effect on future volatility in HK than in China, indicating that HK has been more affected due to its role as an international financial center.
  - (iii) The lower cross-volatility between HK and China than between the United States and China shows that the impact of the United States on China is greater than that of HK in the context of volatility persistence, due mainly to the United States as the origin of the subprime crisis.
  - (iv) The conditional correlation between China and HK has outweighed their correlations with the United States, echoing increasing financial integration between China and HK.

### Contributions to literature and policy relevance
- The paper contributes in three ways:
  - First, it is the first attempt to gauge the impact of the U.S. subprime turmoil on China and HK’s stock markets using a group of univariate and multivariate GARCH models.
  - Second, it compares the different responses of China and HK’s stock markets to the subprime turmoil, reflecting different degrees of financial openness and integration.
  - Third, it can be helpful to authorities in designing policy responses to external shocks, particularly given the background of increasing capital account liberalization in mainland China.

### Stylized facts and preliminary observations (stock market performance)
- By end-October 2008 stock prices in major economies had declined at least 30 percent since their peak in October 2007.
- Specific declines:
  - China experienced a decline of 71 percent (Figure 1).
  - HK experienced a decline of 56 percent (Figure 1).
- U.S. stock market volatility:
  - Volatility had risen since late 2006, with a noticeable spike in mid-2007 and September-October 2008 (Figure 2), coincident with the subprime crisis and increased financial stress after Lehman’s collapse.
- Conditional correlation observation:
  - The conditional correlation between China and HK exceeded their correlations with the United States, indicating increasing financial integration between China and HK.

### Transmission channels from the United States to China and HK
- Loss exposures:
  - Direct exposure of the Chinese financial system to U.S. structured credit products is reported to be about $10 billion.
  - The direct exposure to subprime-related assets accounted for 3.7 percent of total banking assets (CBRC, 2008).
  - Seven Chinese listed banks announced total bond holdings of $721 million in the bankrupt U.S. investment bank Lehman Brothers as of September 22, 2008.
  - HK banks’ total exposure to U.S. subprime securities and structured assets remained well below ½ percent of banking system assets (IMF, 2008a).
- Loss of confidence:
  - Indirect influence through confidence shocks: initial run-up in August 2007 shifted to downside as news about losses and pessimistic projections accumulated, with pronounced reactions to the bankruptcy of Lehman, the sale of Merrill Lynch, and the rescue plan of AIG.
- Uncertain direction of capital flows:
  - Capital flows could be affected in two opposite ways:
    - Credit crunch in advanced economies could drive foreign capital out of China and HK to meet liquidity needs.
    - Some foreign investors might view China and HK as relatively resilient markets and continue or increase investment.
  - Example: In the first two quarters of 2008, China experienced a hot money inflow of about USD 130 billion.

### Paper structure (as organized in the source)
- Section II: Stylized facts regarding stock prices in the United States, China and HK; preliminary discussion of spillover channels and market characteristics.
- Section III: Literature review.
- Section IV: Data properties and methodology.
- Section V: Results of estimated UGARCH and MGARCH models.
- Section VI: Conclusion and policy implications.

*Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09166.pdf*

### 124.9 billion), while in the third and fourth quarter it experienced a hot money

### _wp09166 - 124.9 billion), while in the third and fourth quarter it experienced a hot money

### Market context and recent capital movements
- Reported individual exposures (China’s Securities Daily, Sep 2008): $191.4 million at China Construction Bank (CCB), $151.8 million at the Industrial and Commercial Bank of China (ICBC), $128.8 million at the Bank of China (BOC), $76 million at the CITIC Bank, $70 million at the China Merchants Bank, $70.02 million at the Bank of Communications and $33.6 million at the Industrial Bank.
- Possible additional reported losses due to the loss in Ping An Insurance Company and the inaccuracy of China's financial reporting system.
- Before the subprime crisis: foreign capital inflows driven by expected exchange rate appreciation and asset price rise; subsequently, deleveraging led to capital outflows that outweighed inflows.
- Documented outflow values include an outflow of about USD 7.2 billion and 90 billion (Figure 3) — hot money defined as the residual of foreign exchange reserves minus trade surplus and foreign direct investment.
- U.S. Resident’s Net Foreign Transactions in Foreign Corporate Stocks (proxy for capital flows to China and HK) showed increased volatility in flows to HK since the beginning of the subprime crisis (Figure 4).
- Three market characteristics emphasized:
  - Reform of nontradable shares in 2005 reduced two-tier share-structure distortions and contributed to rising stock prices since 2005.
  - China and HK financial markets differ in openness: China has capital-account restrictions; HK is open to all foreign investors and issuers.
  - Policy choices and financial market development in China have implications for HK’s status as an international financial center and for increasing comovements of stock prices.

### Related theoretical and empirical literature (selected points)
- Theoretical channels: revision of expectations, feedback traders, asymmetric information, herd behavior (e.g., Kodres and Pritsker (1998); Calvo (1999); Calvo and Mendoza (2000)).
- Empirical approaches: correlation/contagion studies (Baig and Goldfajn (1999); Gelos and Sahay (2000); Chan-Lau, Mathieson, and Yao (2004)); VAR models (Guimaraes-Filho et al. (IMF, 2008b); Bayoumi and Swiston (2007)); GARCH and MGARCH approaches (Chan-Lau and Ivaschenko (2003); Booth, Martikainen, and Tse (1997); Worthington and Higgs (2001)); multivariate EVT (Jobst and Kamil (IMF, 2008c)); panel and VAR frameworks (Global Financial Stability Report (IMF, 2008d)).

### Data and methodology
- Data: daily close-of-day stock market price indices based on national currencies for the United States, China, and HK. Returns R_t = LnP_t - LnP_{t-1}. Note: price returns are not corrected for dividends.
- Sample period: January 1, 2005 to October 31, 2008. Subprime turmoil emphasis sample: January 1, 2007–October 31, 2008.
- Domestic control variables: interest rate differential, surprises in M2 growth, CPI inflation, industrial production; surprises in monthly changes of China’s trade balance; difference between HK’s export and import growth used as proxy for HK trade balance.
- Surprise measures: differences between official data on release dates (real-time data) and corresponding market forecasts from Bloomberg and the Consensus Forecast; surprise value used for the release date, 0 for other days in the month.
- Global variables: Chicago Board Options Exchange’s Volatility Index (VIX) and Lehman Brothers Swaption Volatility Index (LBSPX — weighted average of 1-6 month swaption volatilities).
- Econometric framework:
  - UGARCH group: AR(1)-GARCH(1,1) with subprime event indicator (event =1 if subprime negative news, event =0 otherwise). Return equation example: R_t = constant + ζ_t R_{t-1} + λX_{t-1} + φV_t^f + ð event_t + Є_t. Variance: σ_t^2 = α_0 + α_1 Є_{t-1}^2 + β_1 σ_{t-1}^2 + μ event.
  - MGARCH group: Diagonal VECH (identical to Diagonal BEKK) to model conditional cross-moments and identify own- and cross-volatility spillovers. Errors assumed Student's t-distribution due to excess kurtosis.
  - Conditional mean in MGARCH includes own lagged return and foreign lagged returns: R_t = constant + ζ_t R_{t-1} + η R_{f,t-1} + λ X_{t-1} + φ V_t^f + Є_t. Variance: σ_t^2 = α_0 + α_1 Є_{t-1}^2 + β_1 σ_{t-1}^2.

### Diagnostics and preliminary data properties
- Autocorrelation functions implied persistence and AR specification for returns.
- ADF tests: log change in stock price series I(0); VIX and LBSPX I(1).
- Information criteria (Akaike and Schwarz) suggest one lag of dependent variable and regressors.
- Squared returns show persistence and clustering; ARCH tests support GARCH formulation.
- Distribution of squared returns skewed and leptokurtic; standardized residuals show excess kurtosis.

### UGARCH empirical findings — China
- Dependent variables for China: Shanghai Composite Index, Shanghai financial index, FXI (iShare FTSE/Xinhua China 25 index fund; priced in U.S. dollars; includes 11 large financial corporations).
- AR(1)-GARCH(1,1) specifications acceptable for returns and conditional variance.
- Mean equation:
  - Intercept terms for Shanghai Composite Index and financial index insignificant; FXI intercept significant at the 10 percent level with negative sign, indicating FXI declines at a pace of 0.02 percent per day on average, holding other things equal.
  - Lagged daily return significant in FXI model.
  - Interest rate differential significant and negative for Shanghai Composite Index and FXI, indicating monetary tightening negatively affects China’s stock prices.
  - Most other domestic macroeconomic surprises statistically insignificant; surprises had no major impact on China’s stock price returns.
  - Global market volatility variables and subprime events generally statistically insignificant in mean equations for all three models — external shocks did not have significant influence on China’s daily stock prices in these UGARCH specifications.
- Variance equation:
  - Coefficients of ARCH, GARCH and subprime events are positive and statistically significant in all three models, indicating subprime turmoil increased conditional volatility of the financial sector and HK-listed companies’ stock prices.
  - α_1 + β_1 was less than 1 in all cases, producing (positive) finite estimates of unconditional variances.
  - Subprime events variables significant in Shanghai Composite Index and FXI variance models — volatility of the financial sector and FXI equity indices responded to external subprime shocks.

### UGARCH empirical findings — Hong Kong
- Dependent variables for HK: Hang Seng index and Hang Seng financial index.
- Mean equation:
  - Intercept term for all HK models is positive, reflecting HK’s stock prices increase at a pace of 0.01 percent per day on average, holding other things equal.
  - Interest rate differentials statistically significant with positive signs — possibly reflecting capital inflows when HK’s interest rate is higher than the United States.
  - VIX significant in both HK models with negative signs — market volatility negatively impacts HK stock price returns.
  - Subprime event variable significant in the mean equation with negative sign — subprime events materially reduced levels of HK stock price returns.
- Variance equation:
  - Subprime event variable statistically significant with positive sign — subprime events increased conditional volatility of HK stock price returns.
- Overall: subprime events had material negative effects on both conditional volatility and levels of HK stock price returns.

### MGARCH empirical setup and brief result indication
- Two sets of MGARCH models run using composite and financial stock prices for the United States, China, and HK.
- Estimated coefficients for conditional mean return equations using composite indices indicate price spillovers from the United States to HK have been more significant than to China with a coefficient of 0.27 to China and (text truncated in source at this point).
- MGARCH models used to identify sources and magnitudes of spillovers, capturing lagged foreign returns R_{f,t-1} as potential spillover sources.

### Key empirical implications
- Subprime turmoil increased conditional volatility in China’s financial sector and HK-listed companies (significant positive subprime-event coefficients in variance equations).
- HK’s stock market experienced both higher volatility and lower returns associated with subprime events (subprime-event significance in mean and variance).
- China’s daily stock price levels showed limited direct sensitivity to subprime events in UGARCH mean equations, but volatility responses were present, especially for financial and HK-listed segments (FXI).
- Interest rate differentials affected China and HK differently (negative for China, positive for HK), reflecting differing market openness and HK’s role as an international financial center.
- Global market volatility measures (VIX, LBSPX) played a more pronounced negative role for HK returns than for China in UGARCH specifications.

*Italic: Content drawn from the provided IMF working paper excerpt.*

### 0.76 to HK. This may reflect China’s less financial openness than HK.

### _wp09166 - 0.76 to HK. This may reflect China’s less financial openness than HK.

### Methodology
- Multivariate GARCH methodology with conditional variance covariance equations captures volatility and cross-volatility spillovers between the three markets.
- A Diagonal VECH form (A and B restricted to diagonals) is employed; all regressions follow the GARCH (1,1) model and are estimated by maximum likelihood using the Berndt-Hall-Hall-Hausman (BHHH) maximization algorithm.
- Dependent variables include composite indices and financial sector indices (see Table 2). Sample periods include Jan. 1, 2005-Oct. 31, 2008 and Jan. 1, 2007-Oct. 31, 2008.

### Volatility and Spillover Findings
- Own-volatility (ARCH) spillovers:
  - United States: 0.04
  - China: 0.13
  - HK: 0.16
- Cross-volatility effects (composite indices):
  - U.S. ↔ China: 0.07
  - U.S. ↔ HK: 0.08
  - HK ↔ China: 0.15
- Lagged volatility persistence (GARCH term β) for composite indices:
  - United States: 0.97 (Jan. 1, 2007-Oct. 31, 2008 table: Bi1 = 0.9707***/0.9707***)
  - HK: 0.83 (noted as 0.83 in text; Table 10 and related tables report high persistence values)
  - China: insignificant (text)
- Cross-volatility persistence:
  - U.S. → China: 0.40
  - U.S. → HK: 0.90
  - HK ↔ China: 0.37
- Interpretation:
  - Own-volatility spillovers are larger for emerging markets than for developed markets.
  - Higher cross-volatility between China and HK indicates stronger financial integration between China and HK than with the United States.
  - Past volatility shocks in the United States have a more persistent effect on future volatility in HK than in China, reflecting HK’s role as an international financial center.
  - The United States has a slightly larger impact on China than on HK in volatility persistence in the context of the subprime crisis origin.

### Conditional Correlations (Jan. 1, 2007-Oct. 31, 2008)
- Correlations between indices:
  - Cor(LOGSPX,LOGSHCOMP) — correlation between U.S. S&P 500 index and Shanghai Composite index: much lower than with HK (visualized in Figure 7; text statement).
  - Cor(LOGSPX,LOGHSI) — correlation between U.S. S&P 500 index and Hang Seng index: higher than with Shanghai.
  - Cor(LOGSHCOMP,LOGHSI) — correlation between Shanghai Composite index and Hang Seng index: highest among the three markets.
- Interpretation:
  - Conditional correlation analysis indicates limited openness in China.
  - Correlation between China and HK has outweighed correlation with the United States, echoing increasing financial integration between China and HK.

### Robustness Checks
- Models and checks:
  - UGARCH and MGARCH models using financial sector stock prices (Table 7, 8, 11 and 12, Figure 8).
  - Extended starting sample date from January 1, 2007 to January 1, 2005 (Table 7 and 8).
  - Additional dependent and explanatory variables and various changes in estimation methodology.
- Robustness results:
  - Mean and variance equation results generally hold for UGARCH and MGARCH models using financial sector stock prices.
  - Minor differences: coefficients for mean spillover effects and ARCH effects are slightly lower, while GARCH effects are higher than those in the composite index models; China’s own-volatility coefficient becomes significant in some specifications.
  - Overall, results are rather robust with respect to sample period, additional variables, and estimation methodology (Table 9 to 12).

### Policy Conclusions and Implications
- Empirical conclusions:
  - China’s stock market is not immune to the financial crisis; evidenced by price and volatility spillovers from the United States.
  - HK’s equity returns have exhibited more significant price and volatility spillovers from the United States than China’s returns.
  - Past volatility shocks in the United States have a more persistent effect on future volatility in HK than in China.
  - Conditional correlation between China and HK outweighs their correlations with the United States, indicating increasing financial integration between China and HK.
- Policy implications (verbatim positions and recommendations):
  - There is no “decoupling” story for China and HK. No economy can be totally immune to the subprime financial turmoil. Our regression results strongly support this. In our case, even though the Chinese and HK economies may be diversifying their exports and becoming less dependent on the United States, their financial markets are still very much, if not more, influenced by U.S. monetary and financial conditions. China and HK have become increasingly integrated into the global financial system, and the authorities should be alert to the drift toward negative spillover effects.
  - Greater attention should be paid to the risk of a virulent feedback loop between the financial markets and the economy. The deteriorating subprime crisis has increased the downside risks to the United States and global economies, potentially taking their toll on the economies of China and HK.
  - International policy coordination has become more important. The tendency for the markets to transmit volatility rapidly leaves little time for policy-makers to intervene. Recent unprecedented circumstances have called for commensurate action to be taken by central banks, given the rising interdependence among economies. The fact that recent policy actions have been taken by both advanced and emerging market central banks (including China) underlines the need for coordination. Moreover, as evidenced by the increasing volatility spillover between China and HK’s stock markets in our empirical work, the interdependence has been increasing with the strengthening financial integration between the two markets, necessitating further cooperation between the mainland and HK authorities.
  - The Chinese authorities should be more cautious in their approach to capital account liberalization. Given increasing trade openness and financial spillovers from the international market, the capital account is de facto becoming more open over time irrespective of government attempts to control it (Prasad and Rajan, 2008). In this natural opening-up process, it become more urgent for the authorities to strengthen financial markets and relevant infrastructure (governance, the exchange rate system, the supervisory system, etc.) alongside the de facto capital account liberalization. It is argued that in the Asian financial crisis, China suffered less than other Asian countries just because of its lower level of financial openness. And even now, some consider that China’s limited exposure to the subprime crisis is due mainly to its lower level of financial openness. However, in the era of financial globalization, this limited financial openness, which has insulated China well in the past, may be much harder to maintain in the future. The key issue will be how to choose a pragmatic approach to capital account liberalization.

### Key Model Coefficients and Statistics (selected exact values)
- Mean and variance coefficients (selected from Tables 7–12):
  - ARCH term (α) examples:
    - China financial Model (Jan. 1, 2005-Oct. 31, 2008): 0.081**, 0.0962**, 0.0761*** (various specifications)
    - HK (Model examples): 0.1556***, 0.1843***, 0.0952***, 0.1174***
  - GARCH term (β) examples:
    - China financial: 0.6677***, 0.5782***, 0.8661***, 0.9183***, 0.9524***, 0.8579***
    - HK: 0.8068***, 0.7943***, 0.8515***, 0.8327***
  - Subprime events coefficient in variance equations: small positive values, e.g., 0.0001**, 0.0001*** in several models.
- Table 10 (Variance Covariance equations — Jan. 1, 2007-Oct. 31, 2008): selected α and αi coefficients (values preserved):
  - α0: 0.0002 (0.0005)
  - αi1: -0.0525 (0.0428), 0.2461***(0.0423), 0.3353***(0.0259)
  - αi2: 0.0225(0.014), 0.0000(0), 0***(0)
  - αi3: -0.1591***(0.04), -0.0075(0.0525), -0.0510(0.0389)
- Table 9 (Diagonal VECH A and B elements — Jan. 1, 2005-Oct. 31, 2008 and Jan. 1, 2007-Oct. 31, 2008):
  - Ai1 examples: 0.0408***(0.0094), 0.0733***(0.0162), 0.0808***(0.0144)
  - Ai2 examples: 0.0733***(0.0162), 0.1315***(0.0496), 0.145***(0.036)
  - Ai3 examples: 0.0808***(0.0144), 0.145***(0.036), 0.1598***(0.038)
  - Bi1 examples: 0.9707***(0.0073), 0.4028***(0.1424), 0.8987***(0.0182)
  - Bi2 examples: 0.4028***(0.1424), 0.17(0.1182), 0.3729***(0.1364)
  - Bi3 examples: 0.8987***(0.0182), 0.3729***(0.1364), 0.832***(0.0329)

*Source: IMF staff calculations and analysis in the provided chapter (January 1, 2005–October 31, 2008 and January 1, 2007–October 31, 2008 samples).*

### Appendix A. Lists of Subprime Events

### _wp09166 - Appendix A. Lists of Subprime Events

### Timeline of major bankruptcy, writedown, downgrade, and foreclosure events (selected)
- 2006-12-28: Ownit Mortgage Solutions files for bankruptcy.
- 2007-2-12: ResMae Mortgage files for bankruptcy.
- 2007-2-20: Nova Star Financial reports a surprise loss.
- 2007-3-2: Fremont General stops making subprime loans and puts its subprime business up for sale.
- 2007-3-8: New Century Financial, the second largest subprime lender in 2006, stops making loans.
- 2007-3-20: People’s Choice files for bankruptcy.
- 2007-4-2: New Century Financial files for bankruptcy.
- 2007-4-6: American Home Mortgage writes down the value of risky mortgages rated one step above subprime.
- 2007-4-24: National Association of Realtors announces sales of existing homes fell 8.4 percent in March from February.
- 2007-5-9: FOMC minutes: “The correction of the housing sector was likely to continue to weigh heavily on economic activity through most of this year, somewhat longer than previously expected.”
- 2007-5-25: NAR reports sales of existing homes fell by 2.6 percent in April; number of unsold homes reached a record total of 4.2 million.
- 2007-6-6: ZipRealty Inc. announces homes listed for sale in 18 major U.S. metropolitan areas at end of May were up 5.1 percent from April.
- 2007-6-12: RealtyTrac announces U.S. foreclosure filings surged 90 percent in May from May 2006; foreclosure filings were up 19 percent from April.
- 2007-6-14: Goldman Sachs reports flat profit from a year ago due to mortgage market problems.
- 2007-6-22: Bear Stearns pledges up to $3.2 billion to bail out one of its hedge funds because of bad bets on subprime mortgages.
- 2007-7-10: Standard and Poor’s and Moody’s downgrade bonds backed by subprime mortgages; Fitch follows suit.
- 2007-7-18: Bear Stearns announces two hedge funds lost over 90 percent of their value, equal to over $1.4 billion.
- 2007-7-19: The Dow Jones industrials close above 14,000 for the first time.
- 2007-7-25: National Australia Bank tumbles on new $830m provision.
- 2007-7-31: Home prices: 10-City Composite annual decline of 3.4 percent; 20-City Composite annual decline of 2.8 percent.
- 2007-8-1: Two Bear Stearns hedge funds declare bankruptcy; investors file suit against Bear Stearns.
- 2007-8-6: American Home Mortgage files for bankruptcy.
- 2007-8-9: BNP Paribas suspends three funds because of exposure to U.S. mortgages.
- 2007-8-13: Aegis Mortgage files for bankruptcy.
- 2007-8-16: Countrywide Financial draws down $11.5 billion from its credit lines.
- 2007-8-22: RealtyTrac: foreclosures up 93 percent in July 2007 from July 2006; national foreclosure rate in July was one filing for every 693 households; 179,599 filings reported last month, up from 92,845 a year ago.
- 2007-8-27: NAR reports existing home sales declined by 0.2 percent in July, leaving sales 9.0 percent below level 12 months prior.
- 2007-9-5: NAR pending sales: 16.1 percent decline in July from a year ago and a 12.2 percent decline from prior month; July 89.9 level is second lowest in history of the index.
- 2007-9-6: Mortgage Bankers Association quarterly report: delinquency rate for mortgage loans on one-to-four-unit residential properties was 5.12 percent of all loans outstanding in the second quarter of 2007, up 28 basis points from the first quarter of 2007, and up 73 basis points from one year ago. Compared to this time last year, the seriously delinquent rate is 23 basis points higher for prime loans and 304 basis points higher for subprime loans.
- 2007-9-7: BLS: employers cut 4,000 jobs last month, first net decrease since 2003; Dow Jones Industrial Average dropped 200.87 points after release.
- 2007-9-12: David Shulman lowers forecast for housing starts to an annual rate of 1 million to 1.1 million, down from 1.2 million to 1.3 million.
- 2007-9-14: Merrill Lynch signals subprime crisis may hurt third-quarter earnings; reports “fair value adjustments” for potential losses.
- 2007-9-17: Merrill Lynch & Co.'s $1.3 billion bet on subprime lending worsens; Merrill Lynch Bank & Trust Co. lost $111 million through first half of 2007.
- 2007-9-18: RealtyTrac: foreclosure filings surged to 243,000 in August, up 115 percent from August 2006 and 36 percent from July; Impac Mortgage Holdings Inc. says it will quit most lending activities.
- 2007-9-19: Commerce Department: construction of new homes fell by 2.6 percent in August to the slowest pace in 12 years.
- 2007-9-21: HSBC announces plans to close U.S. subprime unit, Decision One Mortgage, and record impairment charge of about $880 million.
- 2007-9-25: NAR: sales of existing single-family homes dropped by 4.3 percent in August compared to July; 10-City Composite annual decline of 4.5 percent.
- 2007-9-27: Luminent Mortgage Capital downgrades its second-quarter profit.
- 2007-10-1: UBS reports its first quarterly loss in nine years; plans to write down $3.4 billion and cut 1,500 jobs in its investment bank.
- 2007-10-3: New York City residential foreclosures hit 698 during third quarter — a 64 percent increase from same period last year. Third-quarter increases: Los Angeles 247 percent, Miami 168 percent. Miami’s foreclosure rate per household is 116 percent higher than Los Angeles and 852 percent higher than New York City.
- 2007-10-4: Moody’s reports subprime mortgage bonds originated in first half of 2007 include loans going delinquent at the fastest recorded rate; predicts 2007 bond delinquencies likely to surpass 2006.
- 2007-10-5: Merrill disclose $5.5bn in subprime loss.
- 2007-10-9: US subprime losses may reach $150 billion: S&P news.
- 2007-10-10: NAR revises down outlook for home sales from 5.92 million to 5.78 million for the year.
- 2007-10-12: Paulson & Co. announces intention to donate $15 million to the Center for Responsible Lending and the National Association of Consumer.
- 2007-10-15: Citigroup acknowledges risk management failures; 57 percent drop in third-quarter profit; wrote off $3.55 billion and set aside $2.24 billion to cover anticipated losses.
- 2007-10-16: NAHB housing market index dropped to 18, lowest since inception in 1985; index declined for eight straight months.
- 2007-10-17: National League of Cities report: 7 out of 10 finance officers from major cities offer pessimistic predictions for economic future.
- 2007-10-18: Standard & Poor’s cuts credit ratings on $23.35 billion of securities backed by pools of home loans offered in first half of year; downgrades hit some AAA securities.
- 2007-10-24: Merrill Lynch writes down $7.9 billion due to exposure to collateralized debt obligations and subprime mortgages; takes a $2.3 billion loss, largest in firm’s history.
- 2007-10-29: John Robbins says approximately a half of million U.S. mortgage borrowers each year for the next few years risk foreclosure.
- 2007-10-30: Shareholders sue Merrill Lynch & Co for issuing false and misleading statements regarding exposure to risky mortgage investments; S&P/Case-Shiller reports housing prices fallen at record rates.
- 2007-11-1: GMAC, Radian post losses; After write-down, Credit Suisse’s profit falls 31 percent.
- 2007-11-2: Banks are expecting $200 billion losses.
- 2007-11-4: Citigroup says it will take additional $8 billion to $11 billion write-down related to subprime mortgages (on top of $5.9 billion reported earlier).
- 2007-11-6: Analyst downgrades Morgan Stanley amid speculation of $6 billion losses due to reduced value of credit investments.
- 2007-11-8: Fed Chairman Ben Bernanke testifies before Joint Economic Committee expressing concern over subprime housing crisis.
- 2007-11-12: Subprime losses may reach $400 billion, analysts expected.
- 2007-11-14: HSBC reports $3.4 billion impairment charge at HSBC Finance Corp; RealtyTrac: foreclosure filings rose in 77 of largest 100 metros; residential foreclosure filings nearly doubled in third quarter from a year earlier.
- 2007-11-15: Barclays Group PLC takes $2.7 billion write-down for losses on securities linked to U.S. subprime mortgage market.
- 2007-11-19: Fannie Mae shares down 7.3 percent to $37.70 on reports it may report a loss between $1 billion to $5 billion on subprime AAA portfolio.
- 2007-11-20: Subprime crisis widely seen spilling into 2009; losses top out at $270 billion.
- 2007-11-21: Shares of Countrywide close below $10 for first time in more than five years.
- 2007-11-22: U.S. subprime losses may hit $300 billion, OECD estimates.
- 2007-11-28: NAR reports sales of existing single-family homes and condominiums dropped by 1.2 percent in October; median price declined to $207,800, a drop of 5.1 percent from October 2006 — single largest one year decline on record.
- 2007-11-29: RealtyTrac: 222,451 foreclosure filings last month; 94 percent increase from October 2006; 2 percent increase from September 2007.
- 2007-12-5: Wall Street Journal reports New York Attorney General Andrew M. Cuomo sent subpoenas to major Wall Street firms to explore role in packaging and selling of subprime mortgages.
- 2007-12-10: UBS announced it would write down an additional $10 billion in subprime losses.
- 2007-12-11: Washington Mutual expected fourth quarter loan losses of $1.6 Billion; expected 3,000 employee layoffs.
- 2007-12-12: Florida Fund reduced by $1.9 billion after SIV losses.
- 2007-12-14: Citigroup rescues SIVs with $58 billion debt bailout; UBS confirms sub-prime $18.4 billion loss.
- 2007-12-18: Commerce Department: housing construction down 3.7 percent for November; 24.2 percent drop in new home construction in 12-month period.
- 2007-12-19: Morgan Stanley announces additional $9.4 billion write-down on subprime-linked investments; selling a $5 billion dollar stake to foreign investment fund.
- 2007-12-20: Bear Stearns announced first quarterly loss in institution’s eight-decade history.
- 2008-1-4: Labor department: unemployment rate rose from 4.7 percent to 5 percent in December.
- 2008-1-10: Countrywide reported late mortgage payments and foreclosures reached highest level ever recorded in December.
- 2008-1-11: Merrill Lynch to write down more than double initial projection related to subprime losses; Bank of America to buy Countrywide Financial.
- 2008-1-15: Citigroup announced mortgage portfolio dropped in value by $18.1 Billion; led to first quarterly loss in 16 years.
- 2008-1-17: Lehman Brothers said it would no longer continue wholesale mortgage lending; announced cut of 1,300 jobs (on top of 2,500 jobs eliminated since June 2007).
- 2008-1-28: New home sales dropped 26.4 percent in 2007; median price of new homes fell by 10.4 percent from December 2006.
- 2008-1-29: Houses in foreclosure rose 79 percent in 2007; December marked fifth straight month with 200,000 or more foreclosure filings.
- 2008-1-30: UBS had worst year of performance in institutional history; S&P announced cutting credit ratings of $534 billion in subprime mortgage backed securities; downgrades could lead to another $265 billion in losses.
- 2008-1-31: Subprime, CDO bank losses may exceed $265 billion.
- 2008-2-1: BLS at Joint Economic Committee: 17,000 jobs lost in January.
- 2008-2-11: Credit writedowns may total $175 billion, analysts expected.
- 2008-2-12: IndyMac posts record loss, calls reserves adequate.
- 2008-2-15: FGIC announced it would seek to split its company into two.
- 2008-2-19: Credit Suisse announced it would write down $1 billion in subprime losses.
- 2008-2-25: January median home price fell and existing home sales dropped for sixth straight month: 0.4 percent drop in sales and 4.6 percent drop in price.
- 2008-2-29: UBS market analysts report subprime losses could reach $600 billion.
- 2008-3-3: Commerce Departments: construction spending plummeted by 1.7 percent in January; HSBC in $17bn credit crisis loss.
- 2008-3-6: Mortgage Bankers Association: by end of 2007, 2.04 percent of all mortgages were in the foreclosure process.
- 2008-3-7: Employers cut 63,000 jobs in February — largest single month decline in almost five years; December and January numbers revised down.
- 2008-3-13: Subprime losses may reach 285 billion dollars: S&P.
- 2008-3-17: Bear Stearns to sell itself to JPMorgan Chase for $2 a share — a 93 percent discount on current stock price.
- 2008-3-18: Wachovia report: Merrill Lynch identified as most at risk major broker behind Bear Stearns due to subprime vulnerability.
- 2008-3-20: Citigroup to cut another 2,000 jobs (on top of 4,200 layoffs announced in January).
- 2008-3-24: JPMorgan Chase raises offer for Bear Stearns to $10 a share from $2/share.
- 2008-3-25: S&P/Case-Shiller index: home prices dropped in January — 19th straight month of declines and largest single-month drop in 20-year history of report.
- 2008-3-31: US Treasury backs regulatory overhaul, broader Fed role.
- 2008-4-1: UBS writes down another $19 billion; Deutsche Bank to write down $4B.
- 2008-4-8: IMF says writedowns on US assets could total $945bn.
- 2008-4-11: China (CBRC) urges caution after subprime crisis.
- 2008-4-14: Wachovia's loss a grim sign for banks.
- 2008-4-15: OECD raises subprime loss tally to $350-420 billion.
- 2008-4-23: Subprime write-downs hit $269 billion.
- 2008-4-29: $800 Billion more in Subprime, Alt-A Mortgages May Head `Underwater'.
- 2008-5-6: Fannie Mae loses $2.2-billion.
- 2008-5-9: Citigroup to wind down $400bn of assets.
- 2008-5-14: Subprime losses by global banks total US$400 billion: Fitch.
- 2008-5-19: Banks keep $35 Billion markdown off income statements.
- 2008-5-20: Credit crisis will extend into 2009; $170 billion in additional writeoffs.
- 2008-5-26: UBS falls after saying more mortgage losses possible.
- 2008-6-2: Morgan Stanley, Merrill, Lehman ratings cut by S&P.
- 2008-6-5: MBIA, Ambac, $1 Trillion of Debt, lose S&P AAA rating.
- 2008-6-9: Lehman Brothers to post $3 billion loss; sets $6 billion stock sale.
- 2008-6-12: KeyCorp to raise $1.5 billion, cut dividend 50 percent.
- 2008-6-18: Paulson & Co. says writedowns may reach $1.3 Trillion.
- 2008-6-25: Countrywide sued by California over mortgage loans.
- 2008-6-30: Florida sues Countrywide.
- 2008-7-7: Freddie Mac, Fannie Mae plunge on capital concerns.
- 2008-7-8: IndyMac falls after regulators say it isn't `well capitalized'.
- 2008-7-10: U.S. mulls future of Fannie, Freddie.
- 2008-7-28: Merrill has $5.7 Billion of writedowns, sells shares.
- 2008-8-6: Freddie Mac posts fourth straight loss, cuts dividend.
- 2008-8-7: “Yes, That's $2 Trillion of Debt-Related Losses.”
- 2008-8-8: Fannie Mae posts fourth straight loss, cuts dividend.
- 2008-8-15: Fannie, Freddie debt faces confidence crisis overseas.
- 2008-8-20: Fannie Mae and Freddie Mac shares fall on fears of size of potential losses.
- 2008-8-26: Housing slump hits BMO's bottom line.
- 2008-9-9: KDB's attempt to invest in Lehman appears over; counterparty credit risks climb on Lehman, WaMu.
- 2008-9-10: Lehman sees $3.9 Billion loss and plans to shed assets.
- 2008-9-15: Lehman files biggest bankruptcy after suitors balk; AIG's ratings cut by S&P, Moody's, threatening fund raising.
- 2008-9-16: WaMu falls as S&P cuts lender's credit rating to junk; Goldman Sachs net plunges; Morgan Stanley profit falls 3 percent.
- 2008-9-17: AIG was on the verge of collapse, and its share plunged 45 percent.
- 2008-9-23: Uncertainty about the bailout plan's prospects.
- 2008-9-24: Uncertainty about the bailout plan's prospects.
- 2008-9-25: Washington Mutual collapsed.
- 2008-9-29: Emergency Economic Stabilization Act defeated 228-205 in the United States House of Representatives; FDIC announces Citigroup Inc. would acquire banking operations of Wachovia.
- 2008-10-1: The financial crisis is spreading to Europe.
- 2008-10-2: Fears of recession overshadowed hopes of rescue plan.
- 2008-10-3: Fears of recession deepened.
- 2008-10-6: Worst week for the stock market in 75 years: Dow Jones lost 22.1 percent that week, down 40.3 percent since October 9, 2007; S&P 500 lost 18.2 percent that week, down 42.5 percent since October 9, 2007.
- 2008-10-7: Bank of America reports third quarter earnings of $1.2 billion; cut dividend by half and planned to raise an additional $10bn in capital.
- 2008-10-8: International Monetary Fund sees major global downturn in bleakest forecast in years.
- 2008-10-9: Worries of worsening credit crisis and global recession.
- 2008-10-10: Japanese company Yamato Life files for bankruptcy — viewed as first direct casualty in Japan from fallout of US subprime mortgage crisis.
- 2008-10-15: Retail sales sank 1.2 percent in September 2008.
- 2008-10-16: Concerns that banks had been rescued too late to stop a slump in the world economy.
- 2008-10-21: ING Groep's senior debt rating cut to Aa3 by Moody's.
- 2008-10-22: Wachovia reports $23.9 billion loss.
- 2008-10-28: Bank of England estimates world credit loss at '£1.8 trillion'.
- 2008-10-30: Commerce Department reported U.S. GDP fell at an annual rate of 0.3 percent in the July-September period, suggesting onset of recession.

### Key quantitative indicators and loss estimates cited in the events
- Foreclosure and delinquency statistics:
  - RealtyTrac: foreclosure filings surged 90 percent in May (2007) from May 2006; up 19 percent from April (2007).
  - RealtyTrac: foreclosures up 93 percent in July 2007 from July 2006; national foreclosure rate in July was one filing for every 693 households; 179,599 filings vs. 92,845 a year ago.
  - RealtyTrac: foreclosure filings surged to 243,000 in August (2007), up 115 percent from August 2006 and 36 percent from July.
  - 2007: houses in foreclosure rose 79 percent.
  - Mortgage Bankers Association: delinquency rate 5.12 percent in Q2 2007; seriously delinquent rate up 304 basis points for subprime loans YoY.
  - By end-2007, 2.04 percent of all mortgages were in the foreclosure process.
- Housing market indicators:
  - Existing-home sales declines: March 8.4 percent MoM; April 2.6 percent MoM; July 0.2 percent MoM and 9.0 percent YoY decline.
  - 10-City Composite annual declines: 3.4 percent (Jul 2007) and 4.5 percent (Aug 2007).
  - New home sales dropped 26.4 percent in 2007.
  - Median new home price down 10.4 percent from December 2006.
  - Median price of a home sold in October declined to $207,800, down 5.1 percent from October 2006.
  - Construction spending plummeted by 1.7 percent in January (2008); housing construction down 3.7 percent in November (2007); 24.2 percent drop in new home construction in 12-month period.
- Corporate writedowns, losses, and capital moves:
  - Bear Stearns pledged up to $3.2 billion to bail out one hedge fund.
  - Bear Stearns hedge funds lost over $1.4 billion (over 90 percent of value).
  - HSBC impairment charge about $880 million (Decision One); HSBC Finance Corp $3.4 billion impairment.
  - UBS planned to write down $3.4 billion; later announced additional $10 billion write-down; wrote down another $19 billion (2008-4-1); confirmed sub-prime $18.4 billion loss (2007-12-14).
  - Countrywide drew down $11.5 billion from credit lines.
  - Merrill Lynch wrote down $5.5bn (2007-10-5) and $7.9 billion (2007-10-24); took $2.3 billion loss on that event; disclosed larger-than-expected additional writedowns.
  - Citigroup wrote off $3.55 billion and set aside $2.24 billion; later announced additional $8 billion to $11 billion write-down.
  - Morgan Stanley announced $9.4 billion additional write-down and selling $5 billion stake (2007-12-19).
  - Citigroup rescues SIVs with $58 billion debt bailout.
  - Washington Mutual expected fourth quarter loan losses of $1.6 Billion and 3,000 layoffs.
  - Lehman planned $6 billion stock sale after posting $3 billion loss.
  - Fannie Mae loses $2.2-billion (2008-5-6).
- Aggregate loss estimates and forecasts:
  - 2007-10-9: US subprime losses may reach $150 billion (S&P).
  - 2007-11-12: Subprime losses may reach $400 billion (analysts expected).
  - 2007-11-20: Losses top out at $270 billion (widely seen).
  - 2007-11-22: U.S. subprime losses may hit $300 billion (OECD estimates).
  - 2008-2-29: UBS analysts report subprime losses could reach $600 billion.
  - 2008-3-13: Subprime losses may reach 285 billion dollars (S&P).
  - 2008-4-8: IMF says writedowns on US assets could total $945bn.
  - 2008-4-15: OECD raises subprime loss tally to $350-420 billion.
  - 2008-4-23: Subprime write-downs hit $269 billion.
  - 2008-5-14: Subprime losses by global banks total US$400 billion (Fitch).
  - 2008-5-20: $170 billion in additional writeoffs expected; credit crisis to extend into 2009.
  - 2008-6-18: Paulson & Co. says writedowns may reach $1.3 Trillion.
  - 2008-8-7: “Yes, That's $2 Trillion of Debt-Related Losses.”
  - 2008-10-28: Bank of England estimates world credit loss at '£1.8 trillion'.

### Market and macroeconomic impacts highlighted
- Equity markets and indices:
  - Dow Jones industrials closed above 14,000 for first time (2007-7-19).
  - Dow Jones dropped 200.87 points after 2007-9-7 jobs report.
  - Week ending 2008-10-6: Dow Jones lost 22.1 percent that week, down 40.3 percent since October 9, 2007; S&P 500 lost 18.2 percent that week, down 42.5 percent since October 9, 2007.
- Employment and real economy:
  - Jobs: employers cut 4,000 jobs (Sep 2007), 17,000 jobs lost in January (2008), employers cut 63,000 jobs in February (2008).
  - Unemployment: rose from 4.7 percent to 5 percent in December (reported 2008-1-4).
  - Retail sales sank 1.2 percent in September 2008.
  - Commerce Department: U.S. GDP fell at an annual rate of 0.3 percent in July-September period (reported 2008-10-30).
- Policy, regulatory, and government responses:
  - US Treasury backs regulatory overhaul, broader Fed role (2008-3-31).
  - Subpoenas issued by New York Attorney General to major Wall Street firms (2007-12-5).
  - Emergency Economic Stabilization Act defeated 228-205 in House (2008-9-29) — FDIC announces Citigroup to acquire Wachovia banking operations.
  - IMF, OECD, Bank of England estimates and warnings about scale of writedowns and global losses across 2008.

### Observations and implied lessons (as reflected across events)
- Rapid escalation from isolated subprime lender failures in 2006-early 2007 to system-wide writedowns, bank capital actions, and sovereign-level loss estimates by 2008.
- Foreclosure rates, delinquency metrics, and home price indices consistently deteriorated across 2007–2008, with repeated monthly and annual declines.
- Loss estimates for subprime and related exposures varied widely across sources and time, ranging from $150 billion to $1.3 Trillion and beyond, reflecting evolving assessments and uncertainty.
- Significant corporate responses included large writedowns, asset sales, capital raises, job cuts, acquisitions (e.g., Bank of America buying Countrywide), forced takeovers, and government/FDIC-facilitated transactions.

*Sources: Bloomberg L.P. and Reuters.*

### Appendix B. Members of FXI US Equity

### Appendix B. Members of FXI US Equity

### Constituents by sector and name
- 1 Non-financial China Mobile Ltd
- 2 Financial Industrial & Commercial Bank of China - H
- 3 Financial China Life Insurance Co-H
- 4 Non-financial PetroChina Co Ltd-H
- 5 Non-financial CNOOC Ltd
- 6 Financial PING AN Insurance Group Co-H
- 7 Financial Bank of China Ltd - H
- 8 Non-financial China Petroleum & Chemical Co-H
- 9 Non-financial China TeleCom CoRP Ltd-H
- 10 Financial China Construction Bank-H
- 11 Financial China Merchants Bank - H
- 12 Financial Bank of Communication Co-H
- 13 Financial Bank of China HONG KONG Holding Ltd
- 14 Non-financial China UNICOM Ltd
- 15 Non-financial China Shenhua Energy Co - H
- 16 Non-financial China Coal Energy Co - H
- 17 Non-financial China Communications Const-H
- 18 Financial China Merchants Holdings International
- 19 Non-financial China NETCOM Group Corp HK Ltd
- 20 Non-financial China COSCO Holdings-H
- 21 Non-financial Huaneng Power International Inc-H
- 22 Non-financial Datang International Power Generating Co-H
- 23 Non-financial Aluminum Corp of China Ltd-H
- 24 Financial China CITIC Bank - H
- 25 Non-financial AIR China Limited-H

### Data provenance
- Source: Bloomberg L.P.

*Appendix B. Members of FXI US Equity — Source: Bloomberg L.P.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09166.pdf_