## 3.         Insurance         DDs

## Source details

**Canonical URL:** [3.         Insurance         DDs](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06296.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06296.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06296.pdf.json)

---

### I. Introduction
- Objective: assess whether increases in real synchronization and advances in financial integration have occurred, and how they relate to systemic risk profiles of financial intermediaries.
- Key conceptual points:
  - Increased synchronization in real activity can reduce cross-country diversification benefits.
  - Financial integration can enhance individual diversification but may increase common exposures and contagion risk across intermediaries.
  - Focus on synchronization of both first and second moments of real activity, integration of equity markets via dispersion of inverse expected discount factors (IEDFs), and systemic risk via distance-to-default (DD) measures for large European banks and insurance companies.
- Data and sample highlights:
  - Monthly seasonally adjusted industrial production growth (IPG) for 10 European countries with complete data for 1961:01-2004:12 (Austria, Belgium, France, Germany, Greece, Italy, Netherlands, Portugal, Sweden, United Kingdom).
  - Equity market IEDF estimation uses 13 sectoral portfolios and 9 countries with Datastream monthly data for 1975:01-2004:12.
  - Portfolios constructed from Datastream indices for banks and insurance companies over 1991.01-2004.12; as of end-2004 the Datastream indices included 63 banks and 53 insurance companies.

### II. Synchronization of Real Activity
- Methodology:
  - Model IPG for country i at date t with an EGARCH(1,1) specification; include a common factor tF measured as a weighted average of IPG rates (weights per Lumsdaine and Prasad (2003)).
  - Cross-sectional variances of conditional means and variances derived and modeled; two measures for 2()(X)tσ: sample cross-sectional variance of IPG and variance of deviations from HP-filter trend (smoothing parameter 129,600).
  - Timing gauged via country-by-country EGARCH estimates with decade dummies.
- Empirical findings (selected coefficients and significance):
  - Common component tF:
    - Mean Equation A1 = -0.00057 (Standard Error 0.00018, T-Statistic -3.22896, p-value 0.00124).
    - Variance Equation B1 = -0.00267 (Standard Error 0.00076, T-Statistic -3.51361, p-value 0.00044).
  - Cross-country dispersion (Table 2):
    - IPG cross-country variance:
      - Mean Equation A1 = -0.007229695 (Std Error 0.004513876, T-Stat -1.60166, Signif 0.10923082).
      - Variance Equation B1 = -0.006901057 (Std Error 0.002589526, T-Stat -2.66499, Signif 0.00769909).
    - De-trended IPG:
      - Mean Equation A1 = -0.006814521 (Std Error 0.001873768, T-Stat -3.63680, Signif 0.00027605).
      - Variance Equation B1 = -0.008062957 (Std Error 0.001681891, T-Stat -4.79398, Signif 0.00000164).
  - Timing: increased synchronization (decline in cross-country variance of IPG rates) began in the early 1980s; convergence in volatility shows more heterogeneous timing across countries (Table 3).
- Interpretation:
  - Synchronization in real activity increased since the early 1980s, mainly driven by business cycle synchronization rather than solely by a decline in magnitude of common shocks.

### III. Equity Markets Integration
- Methodology:
  - Use Flood and Rose (2005) approach to estimate inverse expected discount factors (IEDFs) via a two-step procedure:
    - Step 1: estimate “systematic price” p% for value-weighted industry portfolios using principal components of returns.
    - Step 2: normalize returns by estimated systematic price and obtain IEDF estimates from cross-sectional regressions (Equation (10)).
  - Measure of integration: cross-country variance of estimated IEDFs, 2()(M)tσ; test for declining time trend (Equation (11)).
- Empirical findings:
  - IEDF dynamics well approximated by random walk with drift; tests validate estimation properties (correlated with future returns, first-order autocorrelation not significantly different from zero).
  - Table 4 Panel A: Trend coefficient A1 for cross-country variance of IEDFs is negative and significant:
    - A1 = -0.000240336 (Std Error 0.000106836, T-Stat -2.24957, Signif 0.02447599).
  - Panel B and Panel C (decade dummies): results indicate increased equity market integration approximately since the early 1990s; cross-country variance of decade dummies declines from early 1990s.
- Interpretation:
  - European equity market integration advanced since the early 1990s, consistent with prior literature documenting increasing influence of common European shocks on stock returns and volatility.

### IV. Systemic Risk and Integration
- Measures of systemic risk:
  - Portfolio distance-to-default (DD) for sets of publicly traded, systemically important European banks and insurance companies, based on Black and Scholes (1973) / Merton (1974) structural valuation model.
  - Portfolio DD formula (as given in text): DD = (Ln(PtV/PtL) + (μP t + 0.5 σP^2 * 0.5)) / σP.
  - Portfolio DD is a conservative lower-bound tracker of joint failure probabilities (implicitly higher than weighted sum of individual DDs and underestimates systemic risk due to model assumptions).
  - Monthly DD measures computed using Vassalou and Xing (2004) methodology: implied asset values, returns and volatilities via BSM option valuation using one year of daily equity returns prior to estimation date and accounting liabilities; repeated monthly to construct portfolio DDs.

- Trends in systemic risk (selected findings):
  - Banks (1990-2004):
    - Figure 2 and HP-filter trend: bank systemic risk does not appear to have declined over 1990-2004.
    - Drivers: shifts in banks’ income composition (increase in noninterest income since late 1990s), higher volatility of noninterest income, higher correlation between interest and noninterest income growth (0.79 for the EU-15) reduced diversification benefits; asset return volatility increased, and improvements in capitalization and returns were insufficient to offset volatility increases.
    - Note: “dark side of diversification” evidenced in U.S. and Europe.
  - Insurance:
    - Figure 3: insurance systemic risk dynamics broadly similar to banks in many countries, but with greater cross-country heterogeneity; in some cases a recent decline is observed.

- Convergence of systemic risk measures (Table 5):
  - Modeling approach: model cross-sectional variance of portfolio DDs, 2()(DD)tσ, with EGARCH-type specification including a proxy for common financial-sector risk Yt = average of country financial sector DDs.
  - Banks:
    - Mean Equation A1 = -0.004987358 (Std Error 0.001985143, T-Stat -2.51234, Signif 0.01199330).
    - Variance Equation B1 = -0.017596387 (Std Error 0.005921549, T-Stat -2.97158, Signif 0.00296267).
    - Interpretation: convergence in both mean and variance of systemic risk measures for banks.
  - Insurance:
    - Mean Equation A1 = 0.003887521 (Std Error 0.004944093, T-Stat 0.78630, Signif 0.43169399).
    - Variance Equation B1 = -0.021483407 (Std Error 0.023667905, T-Stat -0.90770, Signif 0.36403563).
    - Interpretation: no evidence of convergence in mean or variance for insurance companies.

- Sensitivity of systemic risk measures to real and financial shocks:
  - Objective: assess whether sensitivity of bank and insurance systemic risk profiles to common real shocks (common component Ft of real activity) and common financial shocks (Datastream European stock market index Rt) has changed.
  - Model for monthly changes in DD: Δρit = αi + β1j * Ft + β2j * Rt + γ * ρi,t-1 + εit; coefficients β allowed to vary across three subperiods: 1991.1-1994.12, 1995.1-1998.12, 1999.1-2004.12.

  - Empirical findings (Tables 6-7, country patterns summarized):
    - Banks (Table 6):
      - Sensitivity to common real shocks increased in all countries except Austria, Greece, Italy, and Norway, where it remained approximately constant. In many countries coefficients turned from negative to positive and significantly so.
      - Sensitivity to financial shocks increased in Austria, France, Germany, Greece, Italy, the Netherlands, and Portugal; decreased in Belgium, Ireland, Norway, Spain, Sweden, and the United Kingdom.
    - Insurance (Table 7):
      - Sensitivity to common real shocks increased in all countries except Austria, Denmark, Italy, and the United Kingdom, where it remained approximately constant.
      - Sensitivity to financial shocks increased in France, Germany, Italy, and the Netherlands; decreased in Denmark, Ireland, and Norway.
  - Summary interpretation:
    - Significant increase in sensitivity of bank systemic risk profiles to common real shocks in virtually all countries (no country documented a decline).
    - Sensitivities to financial shocks increased for institutions in most large countries but declined in some (notably Spain and the United Kingdom for banks).
    - Insurance companies show broadly similar increased exposure to common real shocks, with heterogeneous changes in exposure to common financial shocks; largest-country insurers generally show increased sensitivity to both shock types.

### V. Conclusion and Policy Implications
- Core findings:
  - Increased synchronization of real activity since the early 1980s.
  - Increased equity market integration since the early 1990s (declining cross-country dispersion of IEDFs).
  - Bank systemic risk profiles did not decline during 1990-2004; banks’ DD measures converged in levels and volatility.
  - Sensitivity of bank systemic risk profiles to common real and financial shocks has significantly increased in most countries; insurance firms show increased sensitivity to common real shocks and mixed results for financial shocks.
- Policy implication highlighted by authors:
  - Increased real synchronization and advances in financial integration do not necessarily imply greater financial stability; enhanced monitoring of increased interdependencies in risk profiles among institutions and across markets is an important supervisory task as integration progresses.

### Appendix Table 1. Banks and Insurance Companies — Total Assets, end 2004 (In millions of euro) (selected entries and presentation notes)
- Austria
  - BANK AU.CREDITANSTALT 145,680
  - ERSTE BANK 139,390
  - UNIQA 15,125 (Insurance)
  - WIENER STAEDT VZ 12,525 (Insurance)
  - AMB GENERALI HDG. 85,632 (Insurance)
  - CATTOLICA ASSICURAZIONI 14,773 (Insurance)
  - MILANO ASSIC. 9,804 (Insurance)
  - PREMAFIN-HLDG.DI PART. 31,493 (Insurance)
- Germany
  - BANKGESELLSCHAFT BERLIN 152,041
  - BAYER.HYPO-UND-VBK. 463,496
  - COMMERZBANK 419,674
  - DEUTSCHE BANK 836,368
  - ALLIANZ 958,579 (Insurance)
  - AXA VERSICHERUNG 7,219 (Insurance)
  - DBV-WINTERTHUR HOLDING 22,398 (Insurance)
  - GERLING 4,327 (Insurance)
  - HANNOVER RUCK. 30,615 (Insurance)
  - ERGO VERSICHERUNG 108,988 (Insurance)
  - KOELN.RUCK. 9,390 (Insurance)
  - KOELN.VERWALT.GESELL. 69 (Insurance)
- Italy
  - UNICREDITO ITALIANO 264,791
  - SAN PAOLO IMI 209,627
  - CAPITALIA 131,163
  - BANCA INTESA 273,181
  - BANCA MONTE DEI PASCHI 128,662
  - BCA.NAZ.LAVORO 78,111
  - GENERALI HOLDING 10,430 (Insurance)
  - AMB GENERALI HDG. 85,632 (Insurance)
  - FONDIARIA-SAI 31,381 (Insurance)
  - MILANO ASSIC. 9,804 (Insurance)
  - PREMAFIN-HLDG.DI PART. 31,493 (Insurance)
  - RAS 66,229 (Insurance)
  - UNIPOL 32,201 (Insurance)
- (Additional country entries follow the same format in the source; table mixes bank and insurance company entries; some entries flagged as "Insurance".)
- Notes:
  - Values are presented as "Total Assets, end 2004 (In millions of euro)".
  - The source table contains asterisks and occasional duplicated entries (e.g., "BNC.TOTTA & ACORES 28,824 *"; repeated "KOELN.VERWALT.GESELL. 69"; repeated "AMLIN 2,858").

*Source: _wp06296 - 3.         Insurance         DDs (PDF chapter/section).*

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

### _wp06296 - References

### Tables
- 1. EGARCH Estimates for the Common Components of IPG .................................................................. 16
- 2. EGARCH Estimates for Cross-Country Variances of IPG and De-Trended IPG ..... 17
- 3. Country-by-Country EGARCH Estimates for IPG.................................................... 18
- 4. Dependent Variables: Cross-Country Variance of IEDFs and Country IEDFs......... 19
- 5. Dependent Variable: Cross-Country Variance of Portfolios’ DDs............................ 21
- 6. Dependent Variable: Banks DDs Changes ................................................................ 22
- 7. Dependent Variable: Insurance DDs Changes........................................................... 23

Appendix Tables
- A1. Banks and Insurance Companies ................................................................... 24

### Figures
- 1. EGARCH Estimates of IPG Common Component ................................................... 16
- 2. Bank DDs................................................................................................................... 20

*Source: _wp06296 - References (PDF).*

### 3.         Insurance         DDs........................................................................................

### 3.         Insurance         DDs........................................................................................

### I. Introduction
- Objective: assess whether increases in real synchronization and advances in financial integration have occurred, and how they relate to systemic risk profiles of financial intermediaries.
- Key conceptual points:
  - Increased synchronization in real activity can reduce cross-country diversification benefits.
  - Financial integration can enhance individual diversification but may increase common exposures and contagion risk across intermediaries.
  - Focus on synchronization of both first and second moments (means and volatilities) of real activity, integration of equity markets via dispersion of inverse expected discount factors (IEDFs), and systemic risk via distance-to-default (DD) measures for large European banks and insurance companies.
- Data and sample highlights:
  - Monthly seasonally adjusted industrial production growth (IPG) for 10 European countries with complete data for 1961:01-2004:12 (Austria, Belgium, France, Germany, Greece, Italy, Netherlands, Portugal, Sweden, United Kingdom).
  - Equity market IEDF estimation uses 13 sectoral portfolios and 9 countries with Datastream monthly data for 1975:01-2004:12.
  - Portfolios constructed from Datastream indices for banks and insurance companies over 1991.01-2004.12; as of end-2004 the Datastream indices included 63 banks and 53 insurance companies.

### II. Synchronization of Real Activity
- Methodology:
  - Model IPG for country i at date t with an EGARCH(1,1) specification (Equations (1)-(2)); include a common factor tF measured as a weighted average of IPG rates (weights per Lumsdaine and Prasad (2003)).
  - Cross-sectional variances of conditional means and variances derived and modeled (Equations (3)-(6)); two measures for 2()(X)tσ: sample cross-sectional variance of IPG and variance of deviations from HP-filter trend (smoothing parameter 129,600).
  - Timing gauged via country-by-country EGARCH estimates with decade dummies.
- Empirical findings:
  - Common component tF: estimated EGARCH trend coefficients in mean and variance are negative and highly significant.
    - Table 1: Mean Equation A1 = -0.00057 (Standard Error 0.00018, T-Statistic -3.22896, p-value 0.00124); Variance Equation B1 = -0.00267 (Standard Error 0.00076, T-Statistic -3.51361, p-value 0.00044).
  - Cross-country dispersion:
    - Table 2 (cross-country variance of IPG): Mean Equation A1 = -0.007229695 (Std Error 0.004513876, T-Stat -1.60166, Signif 0.10923082); Variance Equation B1 = -0.006901057 (Std Error 0.002589526, T-Stat -2.66499, Signif 0.00769909).
    - Table 2 (de-trended IPG): Mean Equation A1 = -0.006814521 (Std Error 0.001873768, T-Stat -3.63680, Signif 0.00027605); Variance Equation B1 = -0.008062957 (Std Error 0.001681891, T-Stat -4.79398, Signif 0.00000164).
  - Timing: increased synchronization (decline in cross-country variance of IPG rates) began in the early 1980s; convergence in volatility shows more heterogeneous timing across countries (Table 3).
- Interpretation:
  - Synchronization in real activity increased since the early 1980s, mainly driven by business cycle synchronization rather than solely by a decline in magnitude of common shocks.

### III. Equity Markets Integration
- Methodology:
  - Use Flood and Rose (2005) approach to estimate inverse expected discount factors (IEDFs) via a two-step procedure:
    - Step 1: estimate “systematic price” p% for value-weighted industry portfolios using principal components of returns.
    - Step 2: normalize returns by estimated systematic price and obtain IEDF estimates from cross-sectional regressions (Equation (10)).
  - Measure of integration: cross-country variance of estimated IEDFs, 2()(M)tσ; test for declining time trend (Equation (11)).
- Empirical findings:
  - IEDF dynamics well approximated by random walk with drift; tests validate estimation properties (correlated with future returns, first-order autocorrelation not significantly different from zero).
  - Table 4 Panel A: Trend coefficient A1 for cross-country variance of IEDFs is negative and significant: A1 = -0.000240336 (Std Error 0.000106836, T-Stat -2.24957, Signif 0.02447599).
  - Panel B and Panel C (decade dummies): results indicate increased equity market integration approximately since the early 1990s; cross-country variance of decade dummies declines from early 1990s.
- Interpretation:
  - European equity market integration advanced since the early 1990s, consistent with prior literature documenting increasing influence of common European shocks on stock returns and volatility.

### IV. Systemic Risk and Integration
#### A. Measures of Systemic Risk
- Measure: portfolio distance-to-default (DD) for sets of publicly traded, systemically important European banks and insurance companies, based on Black and Scholes (1973) / Merton (1974) structural valuation model.
  - Portfolio DD formula: DD = (Ln(PtV/PtL) + (μP t + 0.5 σP^2 * 0.5)) / σP  (as given in text; asset aggregation and portfolio mean/variance defined).
  - Portfolio DD is a conservative lower-bound tracker of joint failure probabilities (implicitly higher than weighted sum of individual DDs and underestimates systemic risk due to model assumptions).
- Estimation approach:
  - Monthly DD measures computed using Vassalou and Xing (2004) methodology: implied asset values, returns and volatilities via BSM option valuation using one year of daily equity returns prior to estimation date and accounting liabilities; repeated monthly to construct portfolio DDs.

#### B. Trends in Systemic Risk
- Banks (1990-2004):
  - Figure 2 and HP-filter trend: bank systemic risk does not appear to have declined over 1990-2004.
  - Interpretation: shifts in banks’ income composition (increase in noninterest income since late 1990s), higher volatility of noninterest income, higher correlation between interest and noninterest income growth (0.79 for the EU-15) reduced diversification benefits; asset return volatility increased, and improvements in capitalization and returns were insufficient to offset volatility increases.
  - The “dark side of diversification” noted also in U.S. evidence.
- Insurance:
  - Figure 3: insurance systemic risk dynamics broadly similar to banks in many countries, but with greater cross-country heterogeneity; in some cases a recent decline is observed.

#### C. Convergence of Systemic Risk Measures
- Modeling approach:
  - Model cross-sectional variance of portfolio DDs, 2()(DD)tσ, with EGARCH-type specification including a proxy for common financial-sector risk Yt = average of country financial sector DDs (Equations (12)-(13)).
- Empirical findings (Table 5):
  - Banks: convergence in both mean and variance of systemic risk measures.
    - Mean Equation A1 = -0.004987358 (Std Error 0.001985143, T-Stat -2.51234, Signif 0.01199330).
    - Variance Equation B1 = -0.017596387 (Std Error 0.005921549, T-Stat -2.97158, Signif 0.00296267).
  - Insurance: no evidence of convergence in mean or variance.
    - Mean Equation A1 = 0.003887521 (Std Error 0.004944093, T-Stat 0.78630, Signif 0.43169399).
    - Variance Equation B1 = -0.021483407 (Std Error 0.023667905, T-Stat -0.90770, Signif 0.36403563).
- Interpretation:
  - For banks, systemic risk profiles have converged in levels and volatility, indicating increased risk interdependencies across large European banks; no analogous convergence for insurance companies.

#### D. Sensitivity of Systemic Risk Measures to Real and Financial Shocks
- Objective: assess whether sensitivity of bank and insurance systemic risk profiles to common real shocks (common component Ft of real activity) and common financial shocks (Datastream European stock market index Rt) has changed.
- Model for monthly changes in DD: Δρit = αi + β1j * Ft + β2j * Rt + γ * ρi,t-1 + εit; coefficients β allowed to vary across three subperiods: 1991.1-1994.12, 1995.1-1998.12, 1999.1-2004.12 (Equations (14) and specifications in Tables 6-7).
- Empirical findings:
  - Banks (Table 6):
    - Sensitivity to common real shocks increased in all countries except Austria, Greece, Italy, and Norway, where it remained approximately constant. In many countries coefficients turned from negative to positive and significantly so.
    - Sensitivity to financial shocks varied by country: increased in Austria, France, Germany, Greece, Italy, the Netherlands, and Portugal; decreased in Belgium, Ireland, Norway, Spain, Sweden, and the United Kingdom.
  - Insurance (Table 7):
    - Sensitivity to common real shocks increased in all countries except Austria, Denmark, Italy, and the United Kingdom, where it remained approximately constant.
    - Sensitivity to financial shocks increased in France, Germany, Italy, and the Netherlands; decreased in Denmark, Ireland, and Norway.
- Summary interpretation:
  - Significant increase in sensitivity of bank systemic risk profiles to common real shocks in virtually all countries (no country documented a decline).
  - Sensitivities to financial shocks increased for institutions in most large countries but declined in some (notably Spain and the United Kingdom for banks).
  - Insurance companies show broadly similar increased exposure to common real shocks, with heterogeneous changes in exposure to common financial shocks; largest-country insurers generally show increased sensitivity to both shock types.

### V. Conclusion and Policy Implications
- Core findings:
  - Increased synchronization of real activity since the early 1980s.
  - Increased equity market integration since the early 1990s (declining cross-country dispersion of IEDFs).
  - Bank systemic risk profiles did not decline during 1990-2004; banks’ DD measures converged in levels and volatility.
  - Sensitivity of bank systemic risk profiles to common real and financial shocks has significantly increased in most countries; insurance firms show increased sensitivity to common real shocks and mixed results for financial shocks.
- Policy implication highlighted by authors:
  - Increased real synchronization and advances in financial integration do not necessarily imply greater financial stability; enhanced monitoring of increased interdependencies in risk profiles among institutions and across markets is an important supervisory task as integration progresses.

*Source: IMF working paper content unit _wp06296 - 3.         Insurance         DDs (PDF chapter/section).*

### Appendix Table 1. Banks and Insurance Companies

### Appendix Table 1. Banks and Insurance Companies

### Total Assets, end 2004 (In millions of euro)

- Austria
  - BANK AU.CREDITANSTALT 145,680
  - ERSTE BANK 139,390
  - UNIQA 15,125 (Insurance)
  - WIENER STAEDT VZ 12,525 (Insurance)
  - AMB GENERALI HDG. 85,632 (Insurance)
  - CATTOLICA ASSICURAZIONI 14,773 (Insurance)
  - MILANO ASSIC. 9,804 (Insurance)
  - PREMAFIN-HLDG.DI PART. 31,493 (Insurance)

- Germany
  - BANKGESELLSCHAFT BERLIN 152,041
  - BAYER.HYPO-UND-VBK. 463,496
  - COMMERZBANK 419,674
  - DEUTSCHE BANK 836,368
  - ALLIANZ 958,579 (Insurance)
  - AXA VERSICHERUNG 7,219 (Insurance)
  - DBV-WINTERTHUR HOLDING 22,398 (Insurance)
  - GERLING 4,327 (Insurance)
  - HANNOVER RUCK. 30,615 (Insurance)
  - ERGO VERSICHERUNG 108,988 (Insurance)
  - KOELN.RUCK. 9,390 (Insurance)
  - KOELN.VERWALT.GESELL. 69 (Insurance)
  - KOELN.VERWALT.GESELL. 69 (appears duplicated in source)

- Italy
  - UNICREDITO ITALIANO 264,791 (Banks)
  - SAN PAOLO IMI 209,627 (Banks)
  - CAPITALIA 131,163 (Banks)
  - BANCA INTESA 273,181 (Banks)
  - BANCA MONTE DEI PASCHI 128,662 (Banks)
  - BCA.NAZ.LAVORO 78,111 (Banks)
  - GENERALI HOLDING 10,430 (Insurance)
  - AMB GENERALI HDG. 85,632 (Insurance)
  - FONDIARIA-SAI 31,381 (Insurance)
  - MILANO ASSIC. 9,804 (Insurance)
  - PREMAFIN-HLDG.DI PART. 31,493 (Insurance)
  - RAS 66,229 (Insurance)
  - UNIPOL 32,201 (Insurance)

- Belgium
  - ALMANIJ 259,629 (Banks)
  - DEXIA 349,463 (Banks)
  - FORTIS (BRU) 521,524 (Banks)
  - KBC BKVS.HDG. 225,587 (Banks)
  - AXA KONZERN 38,438 (Insurance)
  - AXA VERSICHERUNG 7,219 (Insurance)
  - GENERALI 252,542 (Insurance)

- Denmark
  - DANSKE BANK 248,949 (Banks)
  - JYSKE BANK 16,687 (Banks)
  - SYDBANK 10,453 (Banks)
  - ALM BRAND 2,897 (Insurance)
  - CODAN 5,358 (Insurance)
  - TOPDANMARK 4,516 (Insurance)

- Netherlands
  - ABN AMRO HOLDING 607,263 (Banks)
  - FORTIS (AMS) 521,524 (Banks)
  - ING GROEP CERTS. 866,201 (Banks)
  - MLP 2,378 (Insurance)
  - MUNCH.RUCK.REGD. 203,501 (Insurance)
  - WUESTENROT & WUERTT. 54,380 (Insurance)

- Greece
  - ALPHA BANK 32,917 (Banks)
  - BANK OF GREECE 32,810 (Banks)
  - BANK OF PIRAEUS 16,591 (Banks)
  - EFG EUROBANK ERGASIAS 31,760 (Banks)
  - EMPORIKI BK.OF GREECE 18,236 (Banks)
  - NAT.BK.OF GREECE 52,877 (Banks)
  - ETHNIKI GREEK GEN IN CO 1,621 (Insurance)

- Norway
  - DNB NOR 86,097 (Banks)
  - SPAREBANKEN ROGALAND 7,152 (Banks)
  - STOREBRAND 20,871 (Insurance)

- France
  - BNP PARIBAS 905,001 (Banks)
  - CREDIT AGRICOLE 814,654 (Banks)
  - SOCIETE GENERALE 600,897 (Banks)
  - APRIL GROUP 419 (Insurance)
  - AGF-ASR.GL.DE FRN. 103,549 (Insurance)
  - EULER HERMES 5,716 (Insurance)
  - SCOR 12,542 (Insurance)

- Portugal
  - BANCO BPI 26,166 (Banks)
  - BNC.ESPR.SANTO (BESCL)R 45,894 (Banks)
  - BNC.TOTTA & ACORES 28,824 * (Banks)
  - BCP R 71,678 (Banks)

- Ireland
  - AXA 440,744 (Insurance)
  - ALLIED IRISH BANKS 102,042 (Banks)
  - ANG.IR.BK. 34,305 (Banks)
  - BANK OF IRELAND 106,431 (Banks)
  - FINAXA 4,973 (Insurance)
  - FBD HOLDINGS 1,394 (Insurance)

- Spain
  - BANCO ESPANOL DE CREDITO 66,257 (Banks)
  - BANCO POPULAR ESPANOL 62,522 (Banks)
  - BANCO SANTANDER CENTRAL HISPANO 569,795 (Banks)
  - BBV ARGENTARIA 306,218 (Banks)
  - CORP.MAPFRE 'R' 18,033 (Insurance)
  - GRUPO CATALANA OCCIDENTE 3,536 (Insurance)

- Sweden
  - FNSPK. A 112,758 (Banks)
  - NORDEA BANK 262,523 (Banks)
  - SEB A 175,698 (Banks)
  - SVENSKA HANDBKN 149,071 (Banks)

- United Kingdom
  - BARCLAYS 735,337 (Banks)
  - HBOS 623,776 (Banks)
  - HSBC HDG. 935,023 (Banks)
  - LLOYDS TSB GP. 394,145 (Banks)
  - RYL.BK.OF SCTL. 821,784 (Banks)
  - STD.CHARTERED 103,769 (Banks)
  - ADMIRAL GROUP 708 (Insurance)
  - ALEA GP.HDG.(BERMUDA) 4,646 (Insurance)
  - AMLIN 2,858 (Insurance)
  - BEAZLEY 1,136 (Insurance)
  - BENFIELD GROUP 6,052 (Insurance)
  - BRIT INSURANCE HOLDINGS 3,563 (Insurance)
  - CATLIN GROUP 2,437 (Insurance)
  - CHAUCER HOLDINGS 1,074 (Insurance)
  - COX IN.HOLDINGS 1,104 (Insurance)
  - DOMESTIC & GENERAL GP. 474 (Insurance)
  - HISCOX 2,269 (Insurance)
  - JARDINE LLOYD THOMPSON 3,171 (Insurance)
  - KILN 707 (Insurance)
  - ROYAL & SUN ALL.IN. 28,279 (Insurance)
  - WELLINGTON UNDERWRITING 1,841 (Insurance)
  - AMLIN 2,858 (Insurance) (appears duplicated in source)

### Notes and presentation features from the table
- Values are presented as "Total Assets, end 2004 (In millions of euro)".
- The table mixes bank and insurance company entries; some entries are flagged explicitly as "Insurance" in the source.
- Asterisks and occasional duplicated entries appear in the source (e.g., "BNC.TOTTA & ACORES 28,824 *"; repeated entries such as "KOELN.VERWALT.GESELL. 69" and "AMLIN 2,858").

*Source: Appendix Table 1. Banks and Insurance Companies (content unit: _wp06296 - Appendix Table 1. Banks and Insurance Companies)*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06296.pdf_