## 1.   Limits on Equity Holdings as Proportion of CI’s Capital

## Source details

**Canonical URL:** [1.   Limits on Equity Holdings as Proportion of CI’s Capital](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2006/_cr06214.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2006/_cr06214.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2006/_cr06214.pdf.json)

---

### A. Introduction — scope and concerns
- Purpose:
  - Assess the risk profile of the nonfinancial equity investments of Spanish credit institutions (CIs), based on a market-risk approach.
- Structure:
  - Overview of current situation and problems arising from CIs’ nonfinancial equity investments.
  - Evolution of nonfinancial equity investments by CIs in Spain since 2002 and implications of institutional changes (Basel II, new accounting standards).
  - Value-at-risk (VaR) analysis, using the Basel II suggested approach, of risk profile and implied capital absorption for nonfinancial equity investments held in the banking book.
  - Assessment that a severe shock to the current portfolio may not be systemically important, but could significantly affect individual CIs’ economic capital and profitability.
  - Recommendations: enhance risk management and surveillance for CIs with significant nonfinancial equity investments; encourage the market-based approach rather than IRB (PD/LGD) for these exposures; enhance regulation to prevent conflicts of interest when CIs act as both lender and shareholder.
- Key conceptual concerns:
  - Perceived higher risk of industrial participations relative to lending (market, liquidity, and other risks).
  - Corporate governance issues and conflicts of interest:
    - Lack of arm’s-length relations when decision-makers in a CI also sit on the board of a participated industrial company.
    - Misguided lending decisions to participated companies when significant capital participation value is at risk.
    - Preferred status of a CI in supplying services to a participated company, possibly distorting prices.
    - Conflicts of interest and informational asymmetries relative to other debt and equity holders.

### B. International and regulatory context
- International regulatory practices:
  - Some countries prohibit banks holding industrial participations (examples listed in source: Chile, Argentina, Colombia, Costa Rica, Bolivia, Ecuador, and Indonesia).
  - Others allow industrial participations but impose quantitative limits as a proportion of CIs’ capital.
  - Some countries prohibit banks from controlling participated companies (examples listed in source: Czech Republic, Japan, Canada, and Malta).
  - The UK has relied on a “moral suasion approach” resulting in negligible holdings by banks.
- Table of limits on equity holdings as proportion of CIs’ capital (In percent) — preserved entries:
  - European Directive: Individual 15, Aggregated 60
  - Spain: 15, 60
  - Italy: 3, 15
  - USA: -, 10
  - Australia: 2.5, 5
  - Jamaica: 10, 50
  - Canada: -, 70
  - Sri Lanka: 10, 30
  - Israel: -, 25
  - Cyprus: 10, 25
  - Mexico 1/: -, 5
  - Source note: 1/ The 5-percent limit is over total deposits and approximately equivalent to a limit of 30 percent over capital.

### C. Industrial participations in Spain — evolution and magnitude
- Historical phases:
  - Post–World War II to late 1970s: CIs active in industrial development due to lack of private capital.
  - Late 1980s–early 1990s: strong participation in privatizations; often became “núcleo duro”.
  - Current phase: industrial participations help offset reductions in interest margins on loans.
- Relative importance and market context:
  - Industrial participations represent a significant proportion of CIs’ regulatory capital in Spain.
  - Average value of industrial participations held by the CIs in the sample was around 8 percent of total stock market capitalization at end-2004; for some individual CIs the ratio was much higher.
  - Figure 1 indicates strong growth of stock market capitalization to GDP 1996–2003, suggesting potential for gradual divestment without substantial impact on equity prices.
  - Market Capitalization to GDP timeframe reported: 1996, 1998, 2000, 2003.
- Evolution of industrial participations relative to regulatory capital (average for sample) — (In percent):
  - Commercial Banks: 1996 40.0, 1997 39.1, 1998 38.2, 1999 47.9, 2000 45.4, 2001 41.7, 2002 43.7, 2003 45.6, 2004 55.4
  - Savings Banks: 1996 42.2, 1997 58.2, 1998 60.2, 1999 57.1, 2000 58.2, 2001 54.3, 2002 47.2, 2003 46.9, 2004 50.9
  - Source: Bank of Spain. Note: Data reported is an average for the commercial and savings banks in the sample used for this analysis.

### D. Distribution by sector and profitability
- Industrial Participations Distribution by Sector, 2004-05 — percent of total portfolio at fair value:
  - Commercial banks (Dec-04 / Mar-05):
    - Basic materials, industry and construction: 6.8 / 7.4
    - Consumer goods: 1.8 / 3.1
    - Energy: 48.7 / 36.3
    - Consumer services: 6.3 / 5.1
    - Financials and real estate: 5.4 / 7.2
    - Technology and telecommunications: 31.0 / 41.0
    - Total: 100.0 / 100.0
  - Savings banks (Dec-04 / Mar-05):
    - Basic materials, industry and construction: 11.9 / 11.8
    - Consumer goods: 2.6 / 2.8
    - Energy: 54.0 / 54.9
    - Consumer services: 5.5 / 5.3
    - Financials and real estate: 9.3 / 9.8
    - Technology and telecommunications: 16.6 / 15.5
    - Total: 100.0 / 100.0
- Reported industrial participation portfolio values:
  - Commercial banks: €22.3 billion in 2004 and €23.2 billion in 2005.
  - Savings banks: €25.8 billion in 2004 and €26.9 billion in 2005.
  - Note: Since early 2005, commercial banks have moved a sizable part of their investment from the energy to the telecommunications sector.
- Importance for profits (Dividends to Total Profits, 1996–2004) — (In percent), annual averages (dividends from both financial and nonfinancial participations):
  - Commercial banks: 1996 23.0, 1997 17.1, 1998 16.4, 1999 14.3, 2000 11.6, 2001 14.0, 2002 13.9, 2003 13.1, 2004 17.0
  - Savings banks: 1996 12.0, 1997 13.9, 1998 20.0, 1999 15.4, 2000 21.1, 2001 20.5, 2002 21.4, 2003 23.6, 2004 28.1
  - Source: Bank of Spain.
- Fair value versus book value:
  - Fair values of industrial participations surpass book values by an average of 30 percent (Figure 2; December 2004).
  - Majority of participations are listed: 82 percent for commercial banks and 84 percent for savings banks — enabling transparent evaluation and potential rapid reduction of exposure.
  - Only a few cases of realized losses relative to original investment; those with largest positions have largest buffers.

### E. Impact of new accounting standards and Basel II
- Changes in accounting treatment under IFRS (as adopted by Spain):
  - Previously (until December 2004): Spanish accounting rules allowed CIs to hold participations below 20 percent of the firm’s capital in the banking book and be evaluated by the equity method.
  - Under IFRS:
    - Equity investments representing more than 20 percent of voting power can be held in the banking book (participaciones) and valued by the equity method (more than 20 percent but less than 50 percent) or consolidated (more than 50 percent).
    - Investments of less than 20 percent of voting power should be held in the banking book (available for sale) and should be fair valued, unless no quoted price in an active market or fair value cannot be reliably measured—in which case amortized cost applies.
    - Significant change: banks can only book dividends coming from equity investments (under the 20 percent threshold), whereas before they could book the proportional part of profits of a participated company (equity method).
- Expected effects of accounting and Basel II:
  - Transition period: a larger portion of equity positions will be moved to the trading book and accounted for at fair value due to the 20 percent threshold.
  - More stringent capital requirements under Basel II will induce transfer of equity investments from the banking book to the trading book.
  - Consequences:
    - Reduced ability of CIs to rely on unrealized capital gains to smooth future losses.
    - Increased volatility of certain balance sheet and profit-and-loss items, making industrial participations less appealing.
    - The full impact on individual CIs and the system will likely take several years to become apparent.

### F. Risk assessment approach and VaR methodology
- Analytical questions:
  - What risks would CIs face if the value of their equity holdings fell or became more volatile (e.g., because of lower economic growth)?
  - Are the conditions that motivated CIs’ equity holdings still in place?
  - What policy actions could a regulator/supervisor take to contain/reduce possible negative impacts of CIs’ involvement in the industrial sector?
- Methodology:
  - Use of value-at-risk (VaR) approach suggested in Basel II to analyze risk profile and implied capital absorption of nonfinancial equity investments held in the banking book.
  - Basel II market-based VaR prescriptions:
    - Banks are required to compute VaR at the 99 percent confidence level over a 10-day holding period (Capital Accord amendment of 1996).
    - For equity holdings kept in the banking book as a permanent investment, Basel II recommends applying a 99.5 percent confidence level on a holding period of three months.
    - Positions to be stressed should be measured at their fair value level or take into account all latent revaluation gains.
  - IRB (PD/LGD) vs market-based (VaR):
    - Basel II allows either a market-based (VaR) approach or an Internal Rating Based (IRB) PD/LGD approach for equity holdings in the banking book.
    - PD and VaR are closely related conceptually; Basel II requires consistency: if IRB is used for equity positions in the banking book, the same approach should be applied to credit risk in the loan portfolio.
    - Supervisors may, for a maximum of ten years, exempt particular equity investments held at the time of the Framework publication from IRB treatment under specified conditions.
- Data and scope:
  - Portfolio composition date: December 31, 2004.
  - Quoted companies: 93.
  - Nonquoted companies: 206.
  - Quoted companies: complete set of individual stock prices available starting from February 2002; foreign-currency quoted participations converted to euro for U.S. dollar, British pound, Brazilian real, and Mexican peso.
  - Nonquoted companies: mapped to six Madrid stock exchange sector indices: Basic Materials, Consumer Services, Consumer Goods, Energy, Financials, Telecommunications; daily index data collected from December 31, 2004 to September 2005; series reconstructed back to May 2001 using constituent firms while maintaining December 2004 index weights.
  - Historical period emphasizes inclusion of the 2002 shocks (Enron and WorldCom aftermath) and September 11, 2001 events; chosen period noted as 2001/02-2005.
  - Analysis perspectives: aggregate portfolio, sector-level exposures, and individual VaR for each of the 19 banks.
  - Reporting metric: VaRs computed in terms of total exposure and reported as a percentage of regulatory capital according to the Basel II definition.

### G. Key VaR findings and stress-test results
- Pre-shock capital:
  - CAR before shocks: 12.71 percent.
- Aggregate/average impacts (sampled credit institutions):
  - Average impact on regulatory capital: -8.2 percent.
  - Average impact on profits: 88.0 percent.
  - Average CAR (weighted by credit institutions' own funds): 11.7 percent.
- Commercial banks:
  - Average impact on regulatory capital: -5.6 percent.
  - Average CAR: 11.9 percent.
  - Average impact on profits: 57.1 percent.
  - Worst impact on profits: 116.1 percent.
- Cajas (savings banks):
  - Average impact on regulatory capital: -13.3 percent.
  - Maximum (worst) impact on regulatory capital: -24.1 percent.
  - Average CAR: 11.2 percent.
  - Worst CAR after stress: 8.2 percent.
  - Average impact on profits: 147.2 percent.
  - Worst impact on profits: 250.4 percent (worst observed for the cajas).
- Aggregate portfolio highlights:
  - Average impact over the entire period on regulatory capital under the Basel II definition for all institutions is 8.2 percent, while the corresponding impact on profits is 88 percent.
  - Worst impact over the same period is equivalent to 24.1 percent of regulatory capital or to 250.4 percent of 2004 profits (observed for the cajas).
- Sector-level results:
  - Sector VaR and corresponding capital absorptions generally higher than total-portfolio VaR due to loss of idiosyncratic diversification when using sector indices.
  - Highest contribution to overall risk comes from the energy and telecommunications sectors.
  - Sector-level series show effects of September 11, 2001 and the 2002 shocks.
- Individual CI evolution:
  - Commercial and savings banks largely invested in energy, telecommunications and financial/construction sectors; responses to shocks show similar patterns across institutions.
  - Savings banks (cajas), given relatively larger portfolio positions, report larger impacts on capital and profits.

### H. Conclusions
- Monitoring and systemic assessment:
  - Risks from current levels of CIs’ industrial participations require a high degree of monitoring.
  - Given current portfolio composition and buffers (fair value premium average of 30 percent), a severe shock may not be systemically important for the Spanish financial system, but could materially affect individual CIs’ economic capital and profitability.
- Profit drivers and vulnerabilities:
  - Profits from industrial participations in recent years depend largely on rapid growth in equity prices since 2002; strategy of offsetting shrinking interest margins with dividends and capital gains has been successful but increases risk.
- Effects of regulatory and accounting change:
  - Adoption of Basel II will be more demanding in terms of capital requirements.
  - New IFRS will entail a one-off recognition of unrealized capital gains, followed by higher volatility in CIs' balance sheets.
  - Combined, these changes will diminish incentives for CIs to hold industrial participations.
- System viability:
  - A sizable shock would probably not impair the viability of the system overall.
  - Nonetheless, current exposures are such that considerable losses could be incurred by some smaller CIs.

### I. Recommendations
- Continue and enhance monitoring:
  - Support authorities’ intention to require adoption of Basel II methodologies for industrial participations held in the banking book as soon as Basel II comes into force.
- Consider additional regulatory measures to reduce industrial participations:
  - Implement the most conservative approaches considered in Basel II for industrial participations.
  - For CIs with larger and more complex industrial participation books, the Bank of Spain (BE), while exercising internal model validation, should recommend adoption of the market-based (VaR) approach which provides more demanding capital requirements than IRB and a more adequate and effective monitoring tool.
  - Acknowledge practical considerations: for comparability with other European countries, the BE may choose the PD/LGD approach at least in an early implementation phase.
- Governance restrictions:
  - Introduce regulations to prevent CI directors or decision-making officers serving on the board of a nonfinancial company from taking part in the CI’s operations vis-à-vis that company.

*Source: IMF staff note prepared by Renzo G. Avesani and Jorge Cayazzo (MFD); Silvia Ramírez (MFD) provided research assistance; based on data provided by the Bank of Spain.*

### 1.   Limits on Equity Holdings as Proportion of CI’s Capital....................................................5

### 1.   Limits on Equity Holdings as Proportion of CI’s Capital

### A. Introduction — scope and concerns
- This note assesses the risk profile of the nonfinancial equity investments of Spanish credit institutions (CIs), based on a market-risk approach.
- Structure:
  - Overview of current situation and problems arising from CIs’ nonfinancial equity investments.
  - Evolution of nonfinancial equity investments by CIs in Spain since 2002 and implications of institutional changes (Basel II, new accounting standards).
  - Value-at-risk (VaR) analysis, using the Basel II suggested approach, of risk profile and implied capital absorption for nonfinancial equity investments held in the banking book.
  - Assessment that a severe shock to the current portfolio may not be systemically important, but could significantly affect individual CIs’ economic capital and profitability.
  - Recommendations: enhance risk management and surveillance for CIs with significant nonfinancial equity investments; encourage the market-based approach rather than IRB (PD/LGD) for these exposures; enhance regulation to prevent conflicts of interest when CIs act as both lender and shareholder.

- Key conceptual concerns:
  - Perceived higher risk of industrial participations relative to lending (market, liquidity, and other risks).
  - Potential corporate governance issues and conflicts of interest, specifically:
    - Lack of arm’s-length relations when decision-makers in a CI also sit on the board of a participated industrial company.
    - Misguided lending decisions to participated companies when significant capital participation value is at risk.
    - Preferred status of a CI in supplying services to a participated company, possibly distorting prices.
    - Conflicts of interest and informational asymmetries relative to other debt and equity holders.

### B. International and regulatory context
- Regulatory practices internationally:
  - Some countries prohibit banks holding industrial participations (examples listed in source: Chile, Argentina, Colombia, Costa Rica, Bolivia, Ecuador, and Indonesia).
  - Others allow industrial participations but impose quantitative limits as a proportion of CIs’ capital.
  - Some countries prohibit banks from controlling participated companies (examples listed in source: Czech Republic, Japan, Canada, and Malta).
  - The UK has relied on a “moral suasion approach” resulting in negligible holdings by banks.

- Table of limits on equity holdings as proportion of CIs’ capital (In percent) — preserved entries:
  - European Directive: Individual 15, Aggregated 60
  - Spain: 15, 60
  - Italy: 3, 15
  - USA: -, 10
  - Australia: 2.5, 5
  - Jamaica: 10, 50
  - Canada: -, 70
  - Sri Lanka: 10, 30
  - Israel: -, 25
  - Cyprus: 10, 25
  - Mexico 1/: -, 5
  - Source note: 1/ The 5-percent limit is over total deposits and approximately equivalent to a limit of 30 percent over capital.

### C. Industrial participations in Spain — evolution and magnitude
- Historical phases of CI–industry relationship:
  - Post–World War II to late 1970s: CIs were active in industrial development due to lack of private capital.
  - Late 1980s–early 1990s: CIs participated strongly in privatizations and often became “núcleo duro” (reference group of majority shareholders).
  - Current phase: industrial participations help offset reductions in interest margins on loans.

- Relative importance:
  - Industrial participations represent a significant proportion of CIs’ regulatory capital in Spain (Table 2).
  - On average, the value of industrial participations held by the CIs in the sample was around 8 percent of total stock market capitalization at end-2004. For some individual CIs, the ratio was much higher.
  - Figure 1 indicates strong growth of stock market capitalization to GDP 1996–2003, suggesting potential for gradual divestment without substantial impact on equity prices.
  - Market Capitalization to GDP timeframe reported: 1996, 1998, 2000, 2003 (figure axes preserved in source).

- Evolution of industrial participations relative to regulatory capital (average for sample) — (In percent):
  - Commercial Banks: 1996 40.0, 1997 39.1, 1998 38.2, 1999 47.9, 2000 45.4, 2001 41.7, 2002 43.7, 2003 45.6, 2004 55.4
  - Savings Banks: 1996 42.2, 1997 58.2, 1998 60.2, 1999 57.1, 2000 58.2, 2001 54.3, 2002 47.2, 2003 46.9, 2004 50.9
  - Source: Bank of Spain. Note: Data reported is an average for the commercial and savings banks in the sample used for this analysis.

### D. Distribution by sector and profitability
- Sector concentration (Industrial Participations Distribution by Sector, 2004-05) — percent of total portfolio for each type of CI reported at fair value:
  - For Commercial banks (Dec-04 / Mar-05):
    - Basic materials, industry and construction: 6.8 / 7.4
    - Consumer goods: 1.8 / 3.1
    - Energy: 48.7 / 36.3
    - Consumer services: 6.3 / 5.1
    - Financials and real estate: 5.4 / 7.2
    - Technology and telecommunications: 31.0 / 41.0
    - Total: 100.0 / 100.0
  - For Savings banks (Dec-04 / Mar-05):
    - Basic materials, industry and construction: 11.9 / 11.8
    - Consumer goods: 2.6 / 2.8
    - Energy: 54.0 / 54.9
    - Consumer services: 5.5 / 5.3
    - Financials and real estate: 9.3 / 9.8
    - Technology and telecommunications: 16.6 / 15.5
    - Total: 100.0 / 100.0
  - Reported industrial participation portfolio values:
    - Commercial banks: €22.3 billion in 2004 and €23.2 billion in 2005.
    - Savings banks: €25.8 billion in 2004 and €26.9 billion in 2005.
  - Note: Since early 2005, commercial banks have moved a sizable part of their investment from the energy to the telecommunications sector.

- Importance for profits (Dividends to Total Profits, 1996–2004) — (In percent), annual averages (dividends from both financial and nonfinancial participations):
  - Commercial banks: 1996 23.0, 1997 17.1, 1998 16.4, 1999 14.3, 2000 11.6, 2001 14.0, 2002 13.9, 2003 13.1, 2004 17.0
  - Savings banks: 1996 12.0, 1997 13.9, 1998 20.0, 1999 15.4, 2000 21.1, 2001 20.5, 2002 21.4, 2003 23.6, 2004 28.1
  - Source: Bank of Spain.

- Fair value versus book value:
  - Fair values of industrial participations surpass book values by an average of 30 percent (Figure 2; December 2004).
  - Majority of participations are listed: 82 percent for commercial banks and 84 percent for savings banks — enabling transparent evaluation and potential rapid reduction of exposure.
  - Only a few cases of realized losses relative to original investment; those with largest positions have largest buffers.

### E. Impact of new accounting standards and Basel II
- Changes in accounting treatment under IFRS (as adopted by Spain):
  - Previously (until December 2004): Spanish accounting rules allowed CIs to hold participations below 20 percent of the firm’s capital in the banking book and be evaluated by the equity method.
  - Under IFRS:
    - Equity investments representing more than 20 percent of voting power can be held in the banking book (participaciones) and valued by the equity method (more than 20 percent but less than 50 percent) or consolidated (more than 50 percent).
    - Investments of less than 20 percent of voting power should be held in the banking book (available for sale) and should be fair valued, unless no quoted price in an active market or fair value cannot be reliably measured—in which case amortized cost applies.
    - Significant change: banks can only book dividends coming from equity investments (under the 20 percent threshold), whereas before they could book the proportional part of profits of a participated company (equity method).

- Expected effects of accounting and Basel II:
  - Transition period: a larger portion of equity positions will be moved to the trading book and accounted for at fair value due to the 20 percent threshold.
  - More stringent capital requirements under Basel II will induce transfer of equity investments from the banking book to the trading book.
  - Consequences:
    - Reduced ability of CIs to rely on unrealized capital gains to smooth future losses.
    - Increased volatility of certain balance sheet and profit-and-loss items, making industrial participations less appealing.
    - The full impact on individual CIs and the system will likely take several years to become apparent.

### F. Risk assessment approach and main questions addressed
- Analytical questions posed:
  - What risks would CIs face if the value of their equity holdings fell or became more volatile (e.g., because of lower economic growth)?
  - Are the conditions that motivated CIs’ equity holdings still in place?
  - What policy actions could a regulator/supervisor take to contain/reduce possible negative impacts of CIs’ involvement in the industrial sector?

- Methodology signposted:
  - Use of value-at-risk (VaR) approach suggested in Basel II to analyze risk profile and implied capital absorption of nonfinancial equity investments held in the banking book.

### G. Summary findings highlighted in this section
- Industrial participations have historically produced positive economic returns for CIs and contributed to capital; ratios vary substantially across CIs:
  - Dividends-to-profits range from 4 percent to 60 percent (as reported in source).
  - Industrial participations-to-capital range from 6 percent to 88 percent (as reported in source).
- The Spanish stock market exhibits high volatility (as characterized by Basel Committee background work cited in source).
- Given current portfolio composition and buffers (fair value premium average of 30 percent), a severe shock may not be systemically important for the Spanish financial system, but could materially affect individual CIs’ economic capital and profitability.
- Recommendations (summarized from introduction and later discussion):
  - Enhance risk management practices and supervisory surveillance for CIs with significant nonfinancial equity investments.
  - Encourage adoption of the market-based (VaR) approach for banking-book equity exposures rather than IRB (PD/LGD).
  - Enhance regulation to prevent conflicts of interest when CIs serve simultaneously as lender and shareholder.

*Source: IMF staff note prepared by Renzo G. Avesani and Jorge Cayazzo (MFD); Silvia Ramírez (MFD) provided research assistance.*

### 18.      The analysis performed here focuses on the industrial participations of 19 CIs,

### _cr06214 - 18.      The analysis performed here focuses on the industrial participations of 19 CIs,

### Overview
- Sample: 19 CIs, representing 80 percent of the total industrial participation exposures of the banking system.
- Objective: Evaluate possible vulnerabilities of CIs from industrial participations by conducting historical VaR analysis on the fair value of their equity portfolios, following the Basel Committee recommendations in the New Capital Accord of 2004.
- Focus: Economic capital (risk-bearing capacity) — the level of capital needed by a CI to withstand financial shocks and continue operating — rather than accounting consequences.

### Basel II recommendations and approaches (summary of Box 1 and Box 2)
- Market-based VaR prescriptions:
  - Banks are required to compute VaR at the 99 percent confidence level over a 10-day holding period (Capital Accord amendment of 1996).
  - For equity holdings kept in the banking book as a permanent investment, Basel II recommends applying a 99.5 percent confidence level on a holding period of three months.
  - Positions to be stressed should be measured at their fair value level or take into account all latent revaluation gains.
- IRB (PD/LGD) vs market-based (VaR) approaches:
  - Basel II allows either a market-based (VaR) approach or an Internal Rating Based (IRB) PD/LGD approach for equity holdings in the banking book.
  - PD and VaR are closely related conceptually (PD is a function of leverage and asset volatility; equity volatility is used as a proxy for asset volatility), so correctly applied approaches should produce similar results.
  - Basel II requires consistency: if IRB is used for equity positions in the banking book, the same approach should be applied to credit risk in the loan portfolio.
- Supervisory note:
  - Supervisors may, for a maximum of ten years, exempt particular equity investments held at the time of the Framework publication from IRB treatment under specified conditions.

### Data and methodology
- Portfolio composition date: December 31, 2004.
  - Quoted companies: 93.
  - Nonquoted companies: 206.
- Quoted companies: complete set of individual stock prices available starting from February 2002; foreign-currency quoted participations converted to euro using exchange rates for U.S. dollar, British pound, Brazilian real, and Mexican peso. For simplicity, few foreign currency positions were assumed unhedged.
- Nonquoted companies: mapped to six Madrid stock exchange sector indices with sectors: Basic Materials, Consumer Services, Consumer Goods, Energy, Financials, Telecommunications. Daily index data collected from December 31, 2004 to September 2005; series reconstructed back to May 2001 using constituent firms while maintaining December 2004 index weights; observed index volatilities used to shock individual nonquoted firms.
- Historical period emphasizes inclusion of the 2002 shocks (Enron and WorldCom aftermath) and September 11, 2001 events; chosen period noted as 2001/02-2005.
- Analysis perspectives:
  - Aggregate portfolio combining all equity exposures.
  - Sector-level exposures.
  - Individual VaR for each of the 19 banks.
- Reporting metric: VaRs computed in terms of total exposure and reported as a percentage of regulatory capital according to the Basel II definition.

### Key VaR findings and stress-test results
- CAR before shocks: 12.71 percent.
- Aggregate/average impacts (sampled credit institutions):
  - Average impact on regulatory capital: -8.2 percent.
  - Average impact on profits: 88.0 percent.
  - Average CAR (weighted by credit institutions' own funds): 11.7 percent.
- Commercial banks:
  - Average impact on regulatory capital: -5.6 percent.
  - Average CAR: 11.9 percent.
  - Average impact on profits: 57.1 percent.
  - Worst impact on profits: 116.1 percent.
- Cajas (savings banks):
  - Average impact on regulatory capital: -13.3 percent.
  - Maximum (worst) impact on regulatory capital: -24.1 percent.
  - Average CAR: 11.2 percent.
  - Worst CAR after stress: 8.2 percent.
  - Average impact on profits: 147.2 percent.
  - Worst impact on profits: 250.4 percent (worst observed for the cajas).
- Aggregate portfolio result highlights:
  - Average impact over the entire period on regulatory capital under the Basel II definition for all institutions is 8.2 percent, while the corresponding impact on profits is 88 percent.
  - Worst impact over the same period is equivalent to 24.1 percent of regulatory capital or to 250.4 percent of 2004 profits (observed for the cajas).
- Sector-level results:
  - Sector VaR and corresponding capital absorptions generally higher than total-portfolio VaR due to loss of idiosyncratic diversification when using sector indices.
  - Highest contribution to overall risk comes from the energy and telecommunications sectors.
  - Sector-level series show effects of September 11, 2001 and the 2002 shocks.
- Individual CI evolution:
  - Commercial and savings banks largely invested in energy, telecommunications and financial/construction sectors; responses to shocks show similar patterns across institutions.
  - Savings banks (cajas), given relatively larger portfolio positions, report larger impacts on capital and profits.

### Conclusions
- Risks from current levels of CIs’ industrial participations require a high degree of monitoring.
- Profits from industrial participations in recent years depend largely on rapid growth in equity prices since 2002; strategy of offsetting shrinking interest margins with dividends and capital gains has been successful but increases risk.
- Adoption of Basel II and new IFRS:
  - Basel II will be more demanding in terms of capital requirements.
  - New IFRS will entail a one-off recognition of unrealized capital gains, followed by higher volatility in CIs' balance sheets.
  - Combined, these changes will diminish incentives for CIs to hold industrial participations.
- System viability:
  - A sizable shock would probably not impair the viability of the system overall.
  - Nonetheless, current exposures are such that considerable losses could be incurred by some smaller CIs.

### Recommendations (as presented)
- Continue and enhance monitoring:
  - Support authorities’ intention to require adoption of Basel II methodologies for industrial participations held in the banking book as soon as Basel II comes into force.
- Consider additional regulatory measures to reduce industrial participations:
  - Implement the most conservative approaches considered in Basel II for industrial participations.
  - For CIs with larger and more complex industrial participation books, the Bank of Spain (BE), while exercising internal model validation, should recommend adoption of the market-based (VaR) approach which provides more demanding capital requirements than IRB and a more adequate and effective monitoring tool.
  - Acknowledge practical considerations: for comparability with other European countries, the BE may choose the PD/LGD approach at least in an early implementation phase.
- Governance restrictions:
  - Introduce regulations to prevent CI directors or decision-making officers serving on the board of a nonfinancial company from taking part in the CI’s operations vis-à-vis that company.

*Source: IMF staff, based on data provided by the Bank of Spain.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2006/_cr06214.pdf_