## _wp11134 — Annex III

## Source details

**Canonical URL:** [_wp11134 — Annex III](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11134.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11134.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11134.pdf.json)

---

### I. Introduction and objective
- Recent sovereign yield behavior:
  - Sovereign CDS spreads: moved from below 20bp in early September 2008 to around 120bp at the height of the crisis (Japan); increased around 600bp in Korea and the Philippines over the same period; rose over 1000bp in Indonesia.
  - Developments often accompanied by sharp currency depreciation, increased exchange rate volatility, or substantial drain in foreign reserves.
- Macroeconomic consequences highlighted:
  - Widespread increase in long-term interest rates affecting consumption and investment.
  - Higher government debt-servicing costs and funding costs, increasing rollover risk and potential real economic losses.
- Objective:
  - Analyze extent to which large fluctuations in Asian sovereign spreads during the crisis reflected changes in global risk aversion versus country-specific risk (worse fundamentals, increased exchange rate risk, or contagion).

### II. Data, sample, and phases
- Sample and period:
  - Time span: from the beginning of 2005 through mid 2010.
  - Sample: eleven Asian sovereign markets.
- Phases identified since mid-2007:
  - Phase I (October 2008–March 2009): spreads tightened markedly and volatility increased following Lehman Brothers collapse; government bond yields moved sharply above swap yield in most emerging economies in the sample.
  - Phase II (April 2009–September 2009): systemic response/normalization phase; financial spillovers were contained and most bond yields moved back closer to the swap yield.
  - Phase III (since October 2009): sovereign debt crisis originating in Europe led to a sharp increase in sovereign spreads in several euro area countries; Asia-Pacific experienced continuation of normalization with spreads near precrisis levels.
- Spread definition and instruments:
  - Spreads measured as sovereign bond yields to the swap rate (fixed rate leg of interest rate swap) of the same maturity and same currency denomination as the bond.
  - Asian sample: 10-year sovereign spreads for Australia, China, Hong Kong, Japan, New Zealand, Philippines, Korea, and Thailand; 5-year sovereign spreads for India, Indonesia, and Malaysia. Swap spreads for Indonesia are included in analysis though not exhibited.

### III. Model design and regressors
- Regression framework: each sovereign spread is regressed on:
  - IGRA (Index of Global Risk Aversion) — global risk aversion.
  - SC (Spillover Coefficient) — contagion / distress dependence.
  - ERRI (Exchange Rate Risk Index) — exchange rate risk.
  - Country-specific fiscal fundamentals: public debt (percent of GDP) and overall balance (percent of GDP).
- Conceptual points:
  - IGRA and SC separate market-wide liquidity/risk-aversion shocks from distress spillovers.
  - ERRI accounts for exchange rate risk in economies with independent currencies and different monetary/exchange-rate regimes.
  - Fiscal variables entered via daily series obtained by linear interpolation of quarterly data to capture low-frequency effects.

### IV. Global Risk Aversion (IGRA)
- Construction and properties:
  - Based on Espinoza and Segoviano (2011).
  - IGRA formula used: IGRA_t = -(1-PoR_t), where PoR_t is the share of the market price of risk in the actual probability of the stress event (as estimated in Espinoza and Segoviano, 2011).
  - IGRA is a single, exogenously given measure for all countries; estimated using the US Libor OIS rate and the VIX; does not rely on country-specific information from the Asian economies.

### V. Contagion and the Spillover Coefficient (SC)
- Purpose: capture distress dependence — the probability of distress of a country conditional on other countries becoming distressed.
- Methodology summary:
  - Extract marginal probabilities P(A_i) and P(A_j) from individual CDS spreads.
  - Obtain joint probability P(A_i, A_j) using CIMDO (nonparametric Kullback cross-entropy approach) to estimate a multivariate empirical distribution.
  - Compute conditional probability P(A_i/A_j) = P(A_i, A_j) / P(A_j).
  - Compute SC(A_i) = ∑ P(A_i/A_j) · P(A_j) for all j ≠ i.
- Assumption: Recovery rate assumed 40 percent for sovereigns (SC is unit-less; scaling by a constant factor does not change results).

### VI. Exchange Rate Risk Index (ERRI) — construction and coefficients
- Components (each measured relative to three months earlier, roughly 65 working days):
  - (i) percentage change in the exchange rate (against the U.S. dollar) relative to its level 3-months earlier;
  - (ii) exchange rate volatility: 3-month rolling averages of the absolute deviation between log(EXR_t) and log(EXR_{t-1}), computed as (1/T) * sum_{j=1}^{T} |log(EXR_j) - log(EXR_{j-1})| with T = 65 working days;
  - (iii) percentage change in foreign reserves relative to three months earlier (monthly reserves linearly interpolated to daily).
- Standardization and aggregation:
  - Each measure transformed via z-score across all countries over January 1, 2005 to June 30, 2010.
  - PCA used; ERRI is the first principal component.
- PCA-derived ERRI coefficients (principal component loadings):
  - appreciation in the exchange rate: 0.63
  - increase in volatility: -0.49
  - increase in foreign reserves: 0.61
- Data-frequency notes:
  - foreign reserve series monthly; daily changes obtained by linear interpolation.
  - Variants using 6-month and 12-month changes produced essentially the same ERRI results as 3-month changes.

### VII. Empirical findings and roles of factors
- General findings:
  - ERRI: deterioration generally leads to a tightening of swap spreads (government bond yields rise relative to swap yields) in most Asian countries in the sample; notable exception is Indonesia.
  - Sharp depreciations and increases in exchange rate volatility after the collapse of Lehman Brothers explain an important share of the increase in bond yields relative to swap rates for many countries.
  - During the systemic outbreak (October 2008–March 2009), exchange rate risk and worsening fiscal fundamentals accounted for a relatively small share of the total increase in bond yields relative to swap yields compared to contagion.
  - During the systemic response phase (April–September 2009), weakening of the dollar against several currencies reduced perceived currency-related risk, contributing to a reversal in ERRI’s adverse effect on spreads.
  - Overall: ERRI’s effect on sovereign spreads—although significant in many cases—was relatively muted compared to the effect of contagion (SC); contagion was the main driver of sovereign spread developments in the Asian sample during the crisis.
- Model estimation context:
  - ERRI included in a GARCH(1,1) specification for sovereign swap spreads alongside IGRA, SC, overall balance (percent of GDP), and debt-to-GDP ratio.
  - Standardization period: January 1, 2005 to June 30, 2010.
  - Estimation via Maximum Likelihood in eViews using the Marquardt Optimization Algorithm.

### VIII. Key numerical and methodological details (selected)
- ERRI measurement specifics:
  - exchange rate change measured relative to 3-month earlier (roughly 65 working days);
  - volatility computed as 3-month rolling average of absolute daily log-change deviations with T = 65 working days;
  - foreign reserves percent change measured relative to three months earlier; monthly reserves interpolated to daily via linear interpolation.
- PCA-derived ERRI coefficients: 0.63, -0.49, 0.61.
- Standardization period: January 1, 2005 to June 30, 2010.

### IX. Tables — summary statistics (selected)
- Distress dependence percentage-contribution tables (Asian sample):
  - Oct 08 - Mar 09: TOT row: 9.2 7.8 8.6 12.4 7.0 8.1 10.3 8.8 11.5 7.4 8.9 100.0; Standard de viation: 1.7
  - Apr 09 - Sep 09: TOT row: 7.5 7.3 6.8 14.6 4.3 9.4 7.3 9.9 12.2 8.8 11.8 100.0; Standard de viation: 2.9
  - Oct 09 - Sep 10: TOT row: 6.9 7.6 7.3 15.8 3.7 9.7 6.7 10.6 12.4 8.6 10.7 100.0; Standard de viation: 3.3
- Distress dependence percentage-contribution tables (Euro area sample):
  - Oct 08 - Mar 09: TOT row: 5.6 7.4 11.4 9.6 12.2 8.5 16.7 8.8 12.3 7.7 100.0; Standard de viation: 3.2
  - Apr 09 - Sep 09: TOT row: 6.1 7.1 11.9 9.9 9.9 10.1 14.0 10.0 13.6 8.3 100.0; Standard de viation: 2.6
  - Oct 09 - Sep 10: TOT row: 1.7 5.4 10.3 12.0 2.1 7.4 2.9 21.8 17.1 19.2 100.0; Standard de viation: 7.4

### X. Conclusions and implications (from Annex III and Box 1)
- The ERRI provides a single indicator combining currency depreciation/appreciation, exchange rate volatility, and changes in foreign reserves to capture exchange rate-related risks relevant for sovereign spreads.
- In the Asian sample analyzed:
  - Sovereign risk spillovers (contagion) were the main driver of sovereign spread developments during the crisis, with ERRI a secondary but meaningful factor.
  - Exchange rate risks amplified spreads during acute depreciation and volatility episodes, but the impact was generally smaller than contagion effects and varied across countries and phases of the crisis.
  - Changes in fiscal fundamentals played a relatively small role overall, though contagion contributions were higher in economies with higher public debt ratios.

*Source: _wp11134 — Annex III (IMF staff calculations and text).*

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

### _wp11134 - References

### Box
- 1. The Exchange Rate Risk Index (ERRI) ......................................................................................... 11

### Figures
- 1. Asian Countries’ Sovereign Swap Spreads ...................................................................................... 6
- 2. Index of Global Risk Aversion (IGRA) ........................................................................................... 7
- 3. Exchange Rate Risk Index (ERRI) ................................................................................................ 10
- 4. Contributions to Distress Dependence: Systemic Outbreak vs. Sovereign Risk Phase ................. 16

### Annexes
- Annex I............................................................................................................................................... 18
- Annex II ............................................................................................................................................. 20

*Source: _wp11134 - References*

### Annex III ..............................................................................................................

### Annex III

### I. INTRODUCTION
- Recent sovereign yield behavior:
  - After the collapse of Lehman Brothers, spreads between government bonds and swaps of the same maturity and currency rose sharply globally, including in Asia.
  - Sovereign CDS spread examples: moved from below 20bp in early September 2008 to around 120bp at the height of the crisis (Japan); increased around 600bp in Korea and the Philippines over the same period; rose over 1000bp in Indonesia.
  - In many cases, these developments were accompanied by a sharp depreciation of the currency, increased exchange rate volatility, or a substantial drain in foreign reserves.
- Macroeconomic consequences of rising sovereign yields:
  - Widespread increase in long-term interest rates affecting consumption and investment.
  - Higher government debt-servicing costs and funding costs, increasing rollover risk and potential real economic losses.
- Potential drivers of sovereign bond market developments during the crisis:
  - Increased global risk aversion and a systemic preference for selected assets by deleveraging, balance sheet-constrained investors.
  - Higher differentiation among sovereigns as some sovereign securities benefited relative to corporate and other riskier assets.
- Objective of the paper:
  - Analyze the extent to which large fluctuations in Asian sovereign spreads during the crisis reflected changes in global risk aversion versus country-specific risk (worse fundamentals, increased exchange rate risk, or contagion).
- Literature context:
  - Prior studies find both liquidity and fundamentals important determinants of emerging market spreads; macro/fiscal fundamentals impact sovereign spreads over certain horizons; financial volatility matters more in the short run.
  - This analysis explicitly models sovereign spreads on measures of global risk aversion, distress dependence, exchange rate risk, and country-specific fiscal fundamentals.
- Methodological novelties:
  - Use of an IGRA (Index of Global Risk Aversion) measure based on Espinoza and Segoviano (2011) that is independent of the country sample.
  - Use of a distress dependence (Spillover Coefficient, SC) measure to separate contagion effects from global liquidity conditions.
  - Introduction of an innovative Exchange Rate Risk Index (ERRI) to capture exchange rate risks for economies with independent currencies and monetary regimes.

### II. DATA AND CONSTRUCTION OF THE MAIN VARIABLES
- Spread definition:
  - Spreads measured as sovereign bond yields to the swap rate (fixed rate leg of interest rate swap) of the same maturity and same currency denomination as the bond.
  - Typical behavior: AAA-rated government bond yields trade below swap yield; lower-rated bonds may trade above swap yield.
- Dataset coverage:
  - Time span: from the beginning of 2005 through mid 2010.
  - Sample: eleven Asian sovereign markets.
  - Phases identified since mid-2007:
    - Phase I (October 2008–March 2009): spreads tightened markedly and volatility increased following Lehman Brothers collapse; in most emerging economies in the sample government bond yields moved sharply above the swap yield.
    - Phase II (April 2009–September 2009): systemic response/normalization phase; financial spillovers were contained and most bond yields moved back closer to the swap yield.
    - Phase III (since October 2009): sovereign debt crisis originating in Europe led to a sharp increase in sovereign spreads in several euro area countries; Asia-Pacific experienced continuation of normalization with spreads near precrisis levels.
  - Note: The Asian sample included 10-year sovereign spreads for Australia, China, Hong Kong, Japan, New Zealand, Philippines, Korea, and Thailand; and 5-year sovereign spreads for India, Indonesia, and Malaysia. Swap spreads for Indonesia are included in the analytical analysis though not exhibited in the figure.
- Regression framework:
  - Each sovereign spread is regressed on: (i) global risk aversion, (ii) contagion, (iii) exchange rate risk, and (iv) country-specific fundamentals.

### A. Global Risk Aversion (IGRA)
- Concept:
  - The IGRA typifies the market price of risk and extracts from asset prices the effects of the price of risk to compute the market’s expectation of the probability of distress.
- Construction:
  - Methodology: based on Espinoza and Segoviano (2011).
  - IGRA formula used in this analysis:
    - IGRA_t = -(1-PoR_t)
      - where PoR_t is the share of the market price of risk in the actual probability of the stress event (as estimated in Espinoza and Segoviano, 2011).
  - IGRA properties:
    - Single, exogenously given measure of global risk aversion for all countries in the sample.
    - Estimated using the US Libor OIS rate and the VIX; does not rely on country-specific information from the Asian economies.

### B. Contagion and the Spillover Coefficient (SC)
- Purpose:
  - Capture distress dependence: probability of distress of a country conditional on other countries in the sample becoming distressed.
- Methodology (Segoviano and Goodhart, 2009; CIMDO; Segoviano, 2006):
  - Step (i): Extract marginal probabilities of default P(A_i) and P(A_j) from individual CDS spreads.
  - Step (ii): Obtain joint probability P(A_i, A_j) using the CIMDO methodology to estimate a multivariate empirical (CIMDO-) distribution that characterizes probabilities of distress and distress dependence across sovereigns.
    - CIMDO is a nonparametric Kullback (1959) cross-entropy based approach that does not impose parametric distributional forms and is constrained to match empirical probabilities of distress observed for each sovereign (from CDS spreads).
  - Step (iii): Compute conditional probability P(A_i/A_j) = P(A_i, A_j) / P(A_j) via Bayes’ law.
  - Step (iv): Compute SC for country A_i using:
    - SC(A_i) = ∑ P(A_i/A_j) · P(A_j) for all j ≠ i
      - SC is the weighted sum of the probability of distress of A_i given a default in each other country, weighted by the probability of each of those events.
- Practical assumptions:
  - Recovery rate assumed: 40 percent for sovereigns (common in the literature). Since SC is a relative, unit-less measure, results are unchanged if scaled by a constant factor.

### C. Exchange Rate Risk (ERRI)
- Rationale:
  - Exchange rate movements during the crisis included sharp appreciations of perceived safe-haven currencies and depreciation/reversals in carry-trade currencies; these movements affect demand for bonds and sovereign risk perceptions (especially where public debt is foreign-denominated or contingent private foreign liabilities exist).
  - Exchange rate volatility and changes in foreign reserves can materially affect perceived sovereign risk even for fixed-rate regimes due to reserve depletion risk and potential abandonment of pegs.
- Construction:
  - ERRI is a single indicator that includes recent currency depreciations/appreciations (against the U.S. Dollar as the common numeraire), exchange rate volatility, and changes in foreign reserves.
  - Compiled using Principal Component Analysis (PCA) using data for the 11 Asian economies in the sample.
  - Interpretation: a fall in the index is perceived as an increase in exchange rate-associated risk (sharp depreciation, increased volatility, fall in foreign reserves, or combination).
- Empirical observations:
  - ERRI captures:
    - Sharp depreciation of KRW, AUD, and NZD at the height of the financial crisis (AUD and NZD partly reflecting carry-trade reversal against the JPY and significant JPY appreciation).
    - Significant increases in exchange rate volatility for AUD, NZD, KRW, and IDR.
    - Important falls in foreign reserves in Malaysia, Korea, and Indonesia contributing negatively to ERRI.
    - Hong Kong exhibited virtually no exchange rate change or volatility (fixed exchange rate), with foreign reserves slightly rising by end-2008 and a persistently positive ERRI value.
- Additional country-specific variables used in regressions:
  - Each country’s stock of public debt and budget balance as a share of GDP.
    - Daily series for fiscal variables obtained by linear interpolation of quarterly data, assuming these variables explain low-frequency movements in swap spreads with almost no impact on high-frequency (daily) variations.

### III. MODEL DESIGN, ESTIMATION, AND FINDINGS (overview from text)
- Model regressors: IGRA (global risk aversion), SC (contagion/distress dependence), ERRI (exchange rate risk), and country-specific fiscal fundamentals (public debt and budget balance as share of GDP).
- Purpose: disentangle the effects of global risk aversion, contagion, exchange rate risk, and country-specific fundamentals on sovereign spreads in a broad range of emerging and advanced Asian economies.
- Key conceptual points emphasized:
  - IGRA and SC allow separation between market-wide liquidity/risk-aversion shocks and distress spillovers across sovereigns.
  - ERRI allows accounting for exchange rate risk in economies with independent currencies and monetary/exchange rate regimes.
  - Inclusion of these measures is critical because sovereign spread movements can be large without noticeable fiscal position changes.

*Source: Annex III of the supplied IMF working paper content.*

### Box 1. The Exchange Rate Risk Index (ERRI)

### Box 1. The Exchange Rate Risk Index (ERRI)

### Motivation and mechanisms
- Exchange rate related risks can affect sovereign spreads through:
  - sharp depreciation leading to valuation changes in public debt, especially when a sizable share is foreign-currency denominated;
  - contingent liabilities from a highly geared private sector with foreign-currency liabilities;
  - increased exchange market volatility raising uncertainty and prompting flight-to-quality, tightening sovereign spreads;
  - in fixed-exchange regimes, depletion of foreign reserves while maintaining a peg as a source of currency risk.

### Construction of the ERRI
- ERRI components:
  - (i) percentage change in the exchange rate (against the U.S. dollar) relative to its level 3-months earlier (roughly 65 days earlier using working days series);
  - (ii) exchange rate volatility measured as 3-month rolling averages of the absolute deviation between the log-value of the exchange rate on a given day and the log-value the day before, computed as (1/T) * sum_{j=1}^{T} |log(EXR_j) - log(EXR_{j-1})|, where EXR_t denotes the exchange rate at time t and T is 65 working days;
  - (iii) percentage change in foreign reserves relative to three months earlier.
- Data and frequency notes:
  - foreign reserve series were available at a monthly frequency only; daily series of changes in foreign reserves were obtained by linearly interpolating the monthly series.
  - variants using 6-month and 12-month changes yielded essentially the same results as 3-month changes.
- Standardization and aggregation:
  - each of the three measures was transformed via z-score across all countries in the sample and the entire period from January 1, 2005 to June 30, 2010.
  - Principal Component Analysis (PCA) was used to estimate relative weights; ERRI is the first principal component (the eigenvector associated with the largest eigenvalue of the correlation matrix of the three standardized variables).
- ERRI coefficients (principal component loadings):
  - appreciation in the exchange rate: 0.63
  - increase in volatility: -0.49
  - increase in foreign reserves: 0.61

### Empirical role of ERRI in sovereign spread analysis
- Model context:
  - ERRI was included as an explanatory variable in a GARCH(1,1) specification for sovereign swap spreads alongside the Index of Global Risk Aversion (IGRA), the Spillover Coefficient (SC) capturing distress dependence, overall balance (percent of GDP), and debt-to-GDP ratio.
  - Sample period and estimation details: standardized variables over January 1, 2005 to June 30, 2010; estimation via Maximum Likelihood in eViews using the Marquardt Optimization Algorithm.
- Key empirical findings regarding ERRI:
  - a deterioration in ERRI generally leads to a tightening of swap spreads (i.e., government bond yields rise relative to swap yields) in most Asian countries in the sample; the notable exception is Indonesia.
  - the sharp depreciations and increases in exchange rate volatility after the collapse of Lehman Brothers explain an important share of the increase in bond yields relative to swap rates for many countries.
  - during the systemic outbreak (October 2008–March 2009) exchange rate risk and worsening fiscal fundamentals accounted for a relatively small share of the total increase in bond yields relative to swap yields compared to contagion; in some cases (e.g., Hong Kong), exchange rate-related measures played favorably for sovereign spreads due to maintained peg and increased reserves.
  - during the systemic response phase (April–September 2009) the weakening of the dollar against several currencies reduced perceived currency-related risk, contributing to a reversal in ERRI’s adverse effect on spreads.
  - overall, ERRI’s effect on sovereign spreads—although significant in many cases—was relatively muted compared to the effect of contagion (SC).

### Numerical and methodological details (key statistics and parameters)
- exchange rate change measured relative to 3-month earlier (roughly 65 working days);
- volatility computed as 3-month rolling average of absolute daily log-change deviations with T = 65 working days;
- foreign reserves percent change measured relative to three months earlier; monthly reserves interpolated to daily via linear interpolation;
- standardization period: January 1, 2005 to June 30, 2010;
- PCA-derived ERRI coefficients: 0.63 (appreciation), -0.49 (volatility), 0.61 (foreign reserves);
- variants using 6-month and 12-month changes produced essentially the same ERRI results as 3-month changes.

### Conclusions and implications
- The ERRI provides a single indicator combining currency depreciation/appreciation, exchange rate volatility, and changes in foreign reserves to capture exchange rate-related risks relevant for sovereign spreads.
- In the Asian sample analyzed:
  - sovereign risk spillovers (contagion) were the main driver of sovereign spread developments during the crisis, with ERRI a secondary but meaningful factor;
  - exchange rate risks amplified spreads during acute depreciation and volatility episodes, but the impact was generally smaller than contagion effects and varied across countries and phases of the crisis;
  - changes in fiscal fundamentals played a relatively small role overall, though contagion contributions were higher in economies with higher public debt ratios.

*Source: IMF staff calculations and text from Box 1. The Exchange Rate Risk Index (ERRI).*

### ANNEX III

### ANNEX III

### Table 1. Percentage Contributions to Distress Dependence among Asian Sovereigns — Oct 08 - Mar 09
- AUS: 8.8 9.4 12.6 7.8 9.6 10.8 10.0 12.4 8.4 10.2 100.0
- CHN: 10.1 11.9 12.6 7.9 8.7 9.9 8.9 12.2 8.2 9.7 100.0
- HKG: 10.0 11.0 12.5 7.9 8.9 9.7 9.2 12.2 8.5 10.1 100.0
- IDS: 9.6 8.3 8.9 7.1 9.2 11.7 11.0 14.4 8.7 11.2 100.0
- JPN: 10.1 8.8 9.6 12.0 9.3 9.9 9.7 12.4 8.5 9.7 100.0
- MLY: 10.7 8.4 9.4 13.5 8.1 12.6 9.1 12.4 7.6 8.4 100.0
- NZL: 9.7 7.7 8.2 13.8 6.9 10.2 11.5 12.4 8.8 10.7 100.0
- PHL: 10.4 8.0 8.9 14.9 7.8 8.4 13.2 12.6 7.4 8.3 100.0
- KOR: 10.1 8.6 9.3 15.5 7.5 8.9 11.2 8.9 8.3 10.3 100.0
- THL: 10.3 8.6 9.7 13.8 8.0 8.2 11.9 8.6 12.3 8.6 100.0
- IND: 10.5 8.6 9.7 15.1 7.7 7.7 12.2 8.2 13.0 7.3 100.0
- TOT: 9.2 7.8 8.6 12.4 7.0 8.1 10.3 8.8 11.5 7.4 8.9 100.0
- Standard de viation: 1.7

### Table 1. Percentage Contributions to Distress Dependence among Asian Sovereigns — Apr 09 - Sep 09
- AUS: 8.6 8.5 13.4 6.4 10.1 9.7 10.3 12.8 8.9 11.2 100.0
- CHN: 8.8 10.3 14.3 4.8 9.8 7.5 9.4 12.9 9.4 12.7 100.0
- HKG: 9.3 11.1 13.4 5.3 9.7 7.9 9.4 12.5 9.4 12.0 100.0
- IDS: 7.3 7.7 6.7 4.2 11.3 7.7 13.0 15.1 10.8 16.1 100.0
- JPN: 10.6 7.9 8.1 12.6 9.6 8.9 8.5 12.3 8.7 10.8 100.0
- MLY: 8.1 7.8 7.2 16.6 4.6 8.3 11.3 13.4 9.8 12.8 100.0
- NZL: 9.8 7.5 7.4 14.3 6.0 10.5 11.5 12.6 9.1 11.3 100.0
- PHL: 8.0 7.2 6.6 18.3 4.4 10.8 8.7 13.5 9.4 13.0 100.0
- KOR: 8.2 8.2 7.4 17.7 4.7 10.7 8.0 11.2 9.9 13.9 100.0
- THL: 7.7 7.9 7.4 16.9 4.5 10.5 7.7 10.4 13.2 13.7 100.0
- IND: 7.4 8.2 7.2 19.4 4.3 10.4 7.3 11.1 14.3 10.5 100.0
- TOT: 7.5 7.3 6.8 14.6 4.3 9.4 7.3 9.9 12.2 8.8 11.8 100.0
- Standard de viation: 2.9

### Table 1. Percentage Contributions to Distress Dependence among Asian Sovereigns — Oct 09 - Sep 10
- AUS: 8.7 8.9 14.4 5.3 10.4 9.2 10.9 13.0 8.8 10.3 100.0
- CHN: 7.9 11.5 15.5 4.1 10.2 6.8 10.3 13.4 9.2 11.2 100.0
- HKG: 8.4 11.9 14.4 4.7 10.1 7.1 10.1 12.9 9.4 11.0 100.0
- IDS: 6.8 8.1 7.3 8.3 11.7 7.2 14.1 15.4 10.7 14.9 100.0
- JPN: 9.7 8.1 9.1 14.7 10.1 8.8 10.6 12.4 8.2 8.5 100.0
- MLY: 7.5 8.1 7.7 17.9 4.0 7.7 12.2 13.7 9.7 11.5 100.0
- NZL: 9.4 7.6 7.7 15.5 4.9 10.7 12.2 12.9 8.9 10.3 100.0
- PHL: 7.3 7.6 7.2 19.9 3.9 11.3 8.0 13.8 9.3 11.7 100.0
- KOR: 7.6 8.5 8.0 19.0 4.0 11.0 7.4 12.0 9.8 12.7 100.0
- THL: 7.1 8.2 8.1 18.4 3.6 10.8 7.1 11.2 13.6 11.9 100.0
- IND: 6.9 8.2 7.7 20.9 3.1 10.6 6.8 11.6 14.5 9.8 100.0
- TOT: 6.9 7.6 7.3 15.8 3.7 9.7 6.7 10.6 12.4 8.6 10.7 100.0
- Standard de viation: 3.3

---

### Table 2. Percentage Contributions to Distress Dependence among Euro Area Sovereigns — Oct 08 - Mar 09
- GER: 9.9 12.0 11.1 13.7 9.4 15.8 8.4 11.1 8.7 100.0
- FRA: 7.7 11.8 9.7 17.4 8.9 18.0 7.8 11.4 7.3 100.0
- ITA: 6.3 8.6 10.8 14.7 8.9 19.2 9.9 13.9 7.8 100.0
- SPA: 6.5 8.6 13.3 14.3 8.5 18.6 9.0 14.1 7.1 100.0
- NET: 6.9 10.1 13.3 11.5 10.6 17.3 8.9 12.3 9.0 100.0
- BEL: 6.1 8.1 11.3 9.2 14.8 19.0 9.4 14.5 7.5 100.0
- AUT: 5.7 7.9 14.1 12.6 11.4 10.6 11.8 14.4 11.5 100.0
- GRE: 5.3 7.0 12.8 10.5 11.0 9.5 18.4 16.1 9.3 100.0
- IRE: 5.4 7.2 13.3 11.6 11.7 10.5 18.2 12.5 9.6 100.0
- POR: 5.8 7.6 11.6 9.0 12.8 8.4 21.0 9.8 13.8 100.0
- TOT: 5.6 7.4 11.4 9.6 12.2 8.5 16.7 8.8 12.3 7.7 100.0
- Standard de viation: 3.2

### Table 2. Percentage Contributions to Distress Dependence among Euro Area Sovereigns — Apr 09 - Sep 09
- GER: 9.6 12.3 10.3 11.7 10.8 13.6 9.7 13.2 8.8 100.0
- FRA: 8.3 12.6 11.3 12.3 11.2 12.8 9.6 12.5 9.4 100.0
- ITA: 6.7 7.9 10.3 11.3 11.3 16.3 11.2 15.8 9.2 100.0
- SPA: 7.1 8.9 13.1 11.6 11.5 15.0 9.9 13.6 9.3 100.0
- NET: 7.4 9.1 13.3 10.8 11.8 14.2 10.2 14.0 9.1 100.0
- BEL: 6.7 8.1 13.0 10.4 11.5 12.5 10.9 15.2 9.3 100.0
- AUT: 6.4 7.0 14.1 10.2 10.5 11.3 12.7 18.3 9.5 100.0
- GRE: 6.1 7.0 13.1 9.1 10.1 11.0 17.2 17.3 9.1 100.0
- IRE: 6.3 6.9 14.0 9.5 10.5 11.7 18.7 13.1 9.2 100.0
- POR: 6.6 8.2 12.8 10.1 10.7 11.2 15.2 10.8 14.4 100.0
- TOT: 6.1 7.1 11.9 9.9 9.9 10.1 14.0 10.0 13.6 8.3 100.0
- Standard de viation: 2.6

### Table 2. Percentage Contributions to Distress Dependence among Euro Area Sovereigns — Oct 09 - Sep 10
- GER: 10.9 10.8 12.5 3.6 9.5 1.4 19.5 13.7 18.1 100.0
- FRA: 3.5 14.4 14.9 5.2 10.3 4.5 16.6 14.3 16.2 100.0
- ITA: 2.0 8.1 15.7 2.5 8.9 2.8 22.4 16.8 20.8 100.0
- SPA: 2.0 7.2 13.7 2.3 8.7 2.9 23.4 18.0 21.8 100.0
- NET: 3.0 13.3 11.4 12.0 12.0 1.1 17.5 12.5 17.2 100.0
- BEL: 2.3 7.7 12.0 13.4 3.5 4.4 20.7 17.6 18.5 100.0
- AUT: 0.9 8.2 9.2 10.9 0.8 10.7 22.8 14.0 22.6 100.0
- GRE: 1.9 4.9 11.8 14.2 2.0 8.1 3.7 26.5 26.9 100.0
- IRE: 1.6 5.1 10.8 13.3 1.8 8.4 2.7 32.3 23.9 100.0
- POR: 1.9 5.3 12.1 14.7 2.2 8.1 4.0 29.9 21.7 100.0
- TOT: 1.7 5.4 10.3 12.0 2.1 7.4 2.9 21.8 17.1 19.2 100.0
- Standard de viation: 7.4

*Source: IMF staff calculations.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11134.pdf_