## _wp09223 - References

## Source details

**Canonical URL:** [_wp09223 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09223.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09223.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09223.pdf.json)

---

### Introduction: Overview of Australian banking sector resilience
- The Australian banking sector entered the financial turmoil in a sound position and has been resilient to the global crisis.
- Banks’ capital ratios are well above the regulatory requirements.
- The major banks’ AA credit ratings have remained unchanged since the crisis unfolded.
- Major banks were able to raise private equity capital amid the global crisis.
- Impaired assets have increased in the past year but remain low by international standards.
- Key vulnerabilities identified:
  - On liabilities: rollover risks on short-term wholesale funding.
  - On assets: exposure to the household sector and possible corporate sector distress.
- Contingent claim analysis (CCA) results indicate potential losses from corporate exposures could amount to as much as 2 percent of total banking sector loans.
- Recommendation: APRA has regularly stress tested the banking system, but it would be advisable to undertake more extreme stress-test scenarios than applied in the past and include Australia’s overseas subsidiaries.

### The global turmoil: Impact on Australian banks
- Limited direct impact on banks’ asset quality so far due to:
  - Small exposure to U.S. and domestic securitized assets and to U.S. investment vehicles holding structured finance products.
  - Heavily domestic-weighted balance sheets, particularly toward low-risk households.
  - Conservative capital adequacy rules by APRA and regular stress testing.
- Liabilities-side concern:
  - Sizable short-term external debt obligations and disrupted access to offshore wholesale markets after the Lehman Brothers collapse in September 2008.
- Policy measures introduced:
  - Wholesale funding guarantees and guarantees on all deposits under a million dollars introduced in October 2008.
  - These measures facilitated continued market access and liquidity.

### Key financial soundness indicators and selected statistics
- Table 1 highlights (Dec-05 to Mar-09 series shown in source):
  - Return on assets (after tax): "1.0", "0.9", "0.9", "0.4", "0.6", "0.7".
  - Return on equity (after tax): "14.7", "17.8", "16.6", "9.9", "14.4", "11.9".
  - Regulatory capital to risk-weighted assets: "10.4", "10.4", "10.2", "10.9", "11.4", "11.4".
  - Tier I capital to risk-weighted assets: "7.6", "7.4", "7.2", "7.6", "8.2", "8.4".
  - Gross impaired assets to total assets: "0.2", "0.2", "0.2", "0.5", "0.8", "1.0".
  - Net impaired assets to equity: "1.8", "1.9", "1.9", "6.3", "8.5", "9.3".
  - Specific provisions to impaired assets: "37.1", "39.1", "39.5", "31.5", "36.3", "38.0".
  - Risk-weighted assets to total assets: "59.3", "57.1", "54.4", "44.9", "43.0", "42.5".
- Table 2 highlights (Mar-09, Sep-08, etc. series for four largest banks):
  - Return on assets: "0.5", "0.6", "0.6", "0.6", "0.8", "1.1", "0.8", "1.0".
  - Return on equity: "10.3", "10.7", "12.7", "11.9", "12.7", "11.9", "14.3", "21.0".
  - Net interest margin: "2.2", "2.0", "2.1", "2.3", "2.0", "2.0", "2.2", "2.1".
  - Tier one capital ratio (Basel II): "8.2", "7.7", "8.3", "7.3", "8.8", "8.2", "8.4", "7.8".
  - Total capital ratio (Basel II): "11.0", "11.1", "12.2", "10.9", "11.4", "11.6", "11.4", "10.8".
  - TCE/total assets: "4.9", "4.9", "4.4", "4.0", "3.6", "3.8", "4.2", "4.0".
  - Past due 90 days plus/total loans: "0.5", "0.3", "0.5", "0.3", "0.4", "0.3", "0.3", "0.2".
  - Gross impaired to total assets: "0.7", "0.4", "0.6", "0.3", "0.4", "0.1", "0.6", "0.2".
  - Net impaired assets to equity: "8.3", "4.2", "7.3", "4.6", "5.3", "1.5", "5.2", "3.6".
  - Specific provision to gross impaired assets: "36.1", "36.9", "32.5", "30.0", "41.8", "40.8", "42.9", "32.6".
  - Total provision to gross impaired assets: "110.4", "198.1", "105.6", "137.9", "131.8", "250.8", "125.6", "167.8".
  - Cash to total assets: "5.3", "5.3", "2.7", "2.8", "2.0", "1.6", "0.7", "0.9".
  - Cash and due from banks to total assets: "6.4", "7.4", "7.2", "9.9", "4.4", "3.0", "4.8", "6.3".
- The aggregate Tier 1 capital of the smaller domestic banks was more than "10 percent" as of March 2009.

### Asset-quality trends and segmentation
- Past due loans plus impaired assets:
  - Four largest banks: around "1 percent" of total assets as of March 2009.
  - All banks: almost "1½ percent" of assets in March 2009.
  - Smaller domestic banks: nearly "3 percent".
- Causes for smaller banks’ deterioration:
  - Relatively large exposures to some lower quality commercial property and higher share of low-doc lending.
- Note: As of September 2008, total commercial property exposures of smaller domestic banks amounted to "$A 33 billion"—about "12 percent" of smaller banks’ total assets—with specific provisions and security held accounting for "97 percent" of impaired commercial property exposures.

### Liquidity and funding structure
- Wholesale funding accounts for about "50 percent" of total funding, of which about "60 percent" is offshore.
- Financial institutions short-term external debt (residual maturity basis) estimated at about "$A 400 billion (35 percent of GDP)" in March 2009.
- Table 3 liabilities (percent of total) series:
  - Deposits: "41.5", "39.3", "42.5", "45.3" (Dec 06, Dec 07, Dec 08, Jun 09).
  - Domestic wholesale funding: "23.3", "26.7", "24.7", "21.1".
  - Offshore wholesale funding: "29.0", "28.2", "28.7", "30.0".
  - Securitization: "6.2", "5.8", "4.1", "3.6".
  - Equity (as a percent of total liabilities) memorandum item: "5.8", "5.6", "6.2", "6.6".
- Banks raised about "$A 140 billion" between December 2008 and early July 2009 after deposit and wholesale funding guarantees were established in October 2008.
- Banks have started to increase medium-term funding in response to liquidity and rollover risks.

### Contingent claim analysis (CCA): corporate defaults and bank losses
- Nonfinancial corporate sector condition:
  - Entered the crisis with moderate leverage and strong balance sheets.
  - Leverage (debt to assets and debt to equity ratios) remained stable and broadly similar to other advanced countries.
  - Profitability improved since the late 1990s; liquidity increased.
  - Market-based indicators show increased corporate solvency risks since 2008, but risks remain manageable.
- Methodology:
  - Contingent claim approach (CCA) used to estimate forward-looking risk indicators (distance to distress, probability of default) combining balance sheet data and share prices.
  - Translation of corporate expected losses into bank sector losses assumes equal exposure across banks (all banks suffer same relative losses).
- CCA findings and quantitative results:
  - Expected corporate default losses could amount to around "2 percent of GDP" based on historical recovery rates of around "40 percent".
  - Banks’ losses could amount to about "2 percent of total March 2009 loans" (Table 4).
- Caveats:
  - Recent high equity market volatility could magnify distance to default and probability of default measures.
  - Bank lending to the nonfinancial corporate sector slowed significantly in 2009, suggesting that leverage has also declined since 2008.
  - Nonfinancial corporate sector raised significant equity capital during Q1 2009—private nonfinancial companies raised over "$A 18 billion" in shares and other equity in Q1 2009, almost twice the amount raised during Q1 2008.

### Quantified CCA excerpt (selected figures)
- Bank Loans to Corporate Sector: "5.9" (in billions of Australian dollars in source presentation).
- Loan-Loss Provisions: "1,651" (in billions of Australian dollars in source presentation).
- Implied Additional NPL/Losses of Total Bank Loans (selected table formula rows): "1.7", "29.1", "1.8".
- Notes:
  - Sources: MKMV-Credit Edge data as of April 29, 2009; Fund staff calculations.
  - The CCA analysis covers listed companies only.
  - For New Zealand, columns B, C, and D report data as of September 2008; for Australia, B, C, and D report data as of March 2009.

### International comparison and contextual findings
- Australian large banks are less leveraged than banks in comparable countries.
- Expected losses from corporate sector distress in Australia are lower relative to many comparator economies in advanced Asia and ASEAN4.
- Confidence-supporting policy measures (deposit and wholesale guarantees) facilitated substantial bond issuance and rollover of short-term debt.

### International comparison: leverage, deposit, and liquidity ratios (Bankscope 2008 sample)
- Sample: 60 large banks using mostly 2008 Bankscope measures.
- Findings:
  - Australian banks have stronger leverage positions than major Canadian banks.
  - Australian banks have weaker deposit and liquidity ratios than major Canadian banks.
  - Compared to the median of large international banks:
    - The four Australian banks are in the upper half of the sample in terms of leverage.
    - The four Australian banks are around the median in terms of deposit and liquidity ratio.
  - Conclusion: Australian banks seem to be among the stronger institutions, roughly in the same place as their Canadian counterparts.
- Selected exact figures (2008, medians/examples):
  - Equity/Total Assets: National Australia Bank 5.7; Australia and New Zealand Banking Group 5.5; Commonwealth Bank of Australia 5.1; Westpac Banking Corporation 4.5.
  - Deposits/Total Assets: Commonwealth Bank of Australia 57.7; National Australia Bank 55.3; Australia and New Zealand Banking Group 46.5; Westpac Banking Corporation 43.1.
  - Liquid Assets/Total: Westpac Banking Corporation 10.0; Commonwealth Bank of Australia 6.2; National Australia Bank 4.6; Australia and New Zealand Banking Group 3.4.

### Market-based indicators and sector co-movement
- Market indicators show strong co-movement among the four major Australian banks.
- Australian and Canadian banks show strong correlation in equity indicators and divergence from other advanced-country groups.
- Charts referenced in source: Bank Equity Prices (March 2007=100); comparative equity price series for largest banks by jurisdiction; Credit Default Swaps (5-year, average largest banks).

### Limits of simple ratios for predicting equity performance
- Liquidity, deposit, and leverage ratios have weak linear correlation with banks’ stock price evolution during the turmoil:
  - Figure 7 (Share Price Change (end-2006 - June 2009) vs Leverage (end-2006)): R^2 = 8E-06, regression y = 0.0003x + 5.6805.
  - Figure 8 (Share Price Change vs Deposit ratio (end-2006)): R^2 = 0.0058, regression y = 0.0338x + 56.138.
  - Figure 9 (Share Price Change vs Liquidity ratio (end-2006)): R^2 = 0.0187, regression y = 0.0605x + 18.347.
- Implication: assessments should emphasize asset quality and additional complex measures (asset quality, supervision and regulation quality, market structure including securitization, competition).

### Cross-country financial soundness indicators (selected 2007 snapshot)
- Assets to Tier 1 capital multiple examples: Sample Average 33.2; Australia 28.8; Canada 26.4; Iceland 185.1.
- Assets to total capital multiple examples: Sample Average 23.2; Australia 19.9; Canada 21.7; Iceland 156.9.
- Impaired loans to total loans examples: Australia 0.3; Canada 0.4; Iceland 6.2.
- Provisions to impaired loans examples: Australia 216.6; Canada 82.0; Iceland 156.7.
- Return on average assets examples: Australia 1.0; Canada 0.8; Iceland 2.1.
- Return on average equity examples: Australia 17.4; Sample Average 16.1.
- Liquid assets to deposits and ST funding examples: Australia 4.1; Canada 15.1; Sample Average 12.0.
- Notes: Table 8 presents these indicators; assets include off-balance sheet items; some figures expressed as multiples, not percent.

### Asset quality shock analysis — methodology
- Data source: Pillar 3 reports under the Basel II framework for the four large Australian banks (ANZ, Commonwealth Bank of Australia, National Australia Bank, Westpac).
- Loan portfolios consolidated into seven risk categories (I–VII).
- PD (probability of default) and LGD (loss given default) used to calculate likely one-year losses by risk category.
- Shock scenarios modeled:
  - Risk-shifting shock:
    - Corporate loans: shift risk categories up by one category.
    - Mortgages: shift risk categories up by one and double the LGD floor to 40 percent.
    - Other loans: shift risk categories up by one; LGD floors not modified.
  - Six-times PD shock:
    - No risk-category shifting, but PDs increased six-fold; mortgages LGD floor of 40 percent.

### Results — risk-shifting and 40 percent mortgage LGD floor
- Expected total losses by category (Table 9 totals, millions of Australian dollars): 181; 833; 2,770; 12,451; 9,506; 11,343; 5,348.
- Aggregate impact (Table 10):
  - Total losses (millions of Australian dollars): 42,432
  - Mortgage losses (millions of Australia dollars): 15,621
  - Corporate losses (millions of Australian dollars): 18,167
  - Provisions (millions of Australian dollars): 15,942
  - Total losses to capital (millions of Australian dollars): 26,490
  - Risk-weighted assets (millions of Australian dollars): 1,152,573
  - Loss as percent of risk-weighted assets: 2.3
  - Implied new total capital adequacy ratio (average of four banks): 9.2
  - Implied minimum new total capital adequacy ratio among the four banks: 8.3
  - Implied new tier 1 capital adequacy ratio (average of four banks): 6.1
  - Implied new tangible common equity to tangible asset ratio: 3.2
- Interpretation:
  - Banks appear resilient to this shock.
  - Even the hardest hit bank’s total capital adequacy ratio remains above the regulatory 8 percent minimum after this shock.
  - Mortgage insurance effects are not included and could reduce mortgage default impacts.

### Results — six-times PD increase and 40 percent mortgage LGD floor
- Expected total losses by category (Table 11 totals, millions of Australian dollars): 287; 2,091; 3,385; 21,529; 15,451; 11,261; 5,348.
- Aggregate impact (Table 12):
  - Total losses (millions of Australian dollars): 59,353
  - Mortgage losses (millions of Australia dollars): 21,796
  - Corporate losses (millions of Australian dollars): 25,855
  - Provisions (millions of Australian dollars): 15,942
  - Total losses to capital (millions of Australian dollars): 43,411
  - Loss as percent of risk-weighted assets: 3.8
  - Implied new total capital adequacy ratio (average of four banks): 7.8
  - Implied minimum new total capital adequacy ratio among the four banks: 6.6
  - Implied new tier 1 capital adequacy ratio (average of four banks): 4.6
  - Implied new tangible common equity to tangible asset ratio: 2.5
- Interpretation:
  - A six-fold increase in PDs would be needed to reduce the total average capital adequacy ratio below 8 percent.
  - In this scenario average Tier 1 capital would remain above the regulatory minimum of 4 percent, and TCE would fall to 2½ percent.
  - Two banks’ total capital adequacy ratios would shrink below the 8 percent regulatory minimum.

### Comparison with contingent claim analysis (CCA)
- Corporate loan losses (before provisioning for losses), comparisons:
  - Risk-shifting and LGD floor shock: 18.2 (In billions of Australian dollars), 1.1 (Percent of Banking Sector Loans).
  - Six-times PD increase and LGD floor shock: 25.9, 1.6.
  - Contingent claim based analysis: 29.1, 1.8.
- Interpretation: The CCA based estimate of corporate losses is close to the corporate loan loss impact of the more severe shock (six-times PD increase).

### Summary assessment and recommendations
- The modeled shocks do not constitute a rigorous stress test and the results are only indicative of the health of the banking sector.
- APRA has regularly stress tested the banking sector but it would be advisable to:
  - Undertake more extreme scenarios than applied in the past.
  - Include Australia’s overseas subsidiaries in stress testing.
  - Include a more protracted and serious macroeconomic downturn than what was applied in the 2006 Financial Stability Assessment Program.
- Methodological recommendations:
  - Use CCA/BSM measures (one-year-ahead default probabilities and distance-to-default) and incorporate market-implied measures (EICDS spreads) to quantify expected corporate-sector losses and translate them into bank-level expected losses given exposure profiles.

*Source: _wp09223 - References*

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

### _wp09223 - References

### Introduction: Overview of Australian banking sector resilience
- The Australian banking sector entered the financial turmoil in a sound position and has been resilient to the global crisis.
- Banks’ capital ratios are well above the regulatory requirements.
- The major banks’ AA credit ratings have remained unchanged since the crisis unfolded.
- Major banks were able to raise private equity capital amid the global crisis.
- Impaired assets have increased in the past year but remain low by international standards.
- Key vulnerabilities identified:
  - On liabilities: rollover risks on short-term wholesale funding.
  - On assets: exposure to the household sector and possible corporate sector distress.
- Contingent claim analysis (CCA) results indicate potential losses from corporate exposures could amount to as much as 2 percent of total banking sector loans.
- Recommendation: APRA has regularly stress tested the banking system, but it would be advisable to undertake more extreme stress-test scenarios than applied in the past and include Australia’s overseas subsidiaries.

### The global turmoil: Impact on Australian banks
- Limited direct impact on banks’ asset quality so far due to:
  - Small exposure to U.S. and domestic securitized assets and to U.S. investment vehicles holding structured finance products.
  - Heavily domestic-weighted balance sheets, particularly toward low-risk households.
  - Conservative capital adequacy rules by APRA and regular stress testing.
- Liabilities-side concern:
  - Sizable short-term external debt obligations and disrupted access to offshore wholesale markets after the Lehman Brothers collapse in September 2008.
- Policy measures introduced:
  - Wholesale funding guarantees and guarantees on all deposits under a million dollars introduced in October 2008.
  - These measures facilitated continued market access and liquidity.

Key financial soundness indicators and selected statistics (as reported)
- Table 1 highlights (Dec-05 to Mar-09 series shown in source):
  - Return on assets (after tax): "1.0", "0.9", "0.9", "0.4", "0.6", "0.7" (quarterly series reported).
  - Return on equity (after tax): "14.7", "17.8", "16.6", "9.9", "14.4", "11.9".
  - Regulatory capital to risk-weighted assets: "10.4", "10.4", "10.2", "10.9", "11.4", "11.4".
  - Tier I capital to risk-weighted assets: "7.6", "7.4", "7.2", "7.6", "8.2", "8.4".
  - Gross impaired assets to total assets: "0.2", "0.2", "0.2", "0.5", "0.8", "1.0".
  - Net impaired assets to equity: "1.8", "1.9", "1.9", "6.3", "8.5", "9.3".
  - Specific provisions to impaired assets: "37.1", "39.1", "39.5", "31.5", "36.3", "38.0".
  - Risk-weighted assets to total assets: "59.3", "57.1", "54.4", "44.9", "43.0", "42.5".
- Table 2 highlights (Mar-09, Sep-08, etc. series for four largest banks):
  - Return on assets: "0.5", "0.6", "0.6", "0.6", "0.8", "1.1", "0.8", "1.0".
  - Return on equity: "10.3", "10.7", "12.7", "11.9", "12.7", "11.9", "14.3", "21.0".
  - Net interest margin: "2.2", "2.0", "2.1", "2.3", "2.0", "2.0", "2.2", "2.1".
  - Tier one capital ratio (Basel II): "8.2", "7.7", "8.3", "7.3", "8.8", "8.2", "8.4", "7.8".
  - Total capital ratio (Basel II): "11.0", "11.1", "12.2", "10.9", "11.4", "11.6", "11.4", "10.8".
  - TCE/total assets: "4.9", "4.9", "4.4", "4.0", "3.6", "3.8", "4.2", "4.0".
  - Past due 90 days plus/total loans: "0.5", "0.3", "0.5", "0.3", "0.4", "0.3", "0.3", "0.2".
  - Gross impaired to total assets: "0.7", "0.4", "0.6", "0.3", "0.4", "0.1", "0.6", "0.2".
  - Net impaired assets to equity: "8.3", "4.2", "7.3", "4.6", "5.3", "1.5", "5.2", "3.6".
  - Specific provision to gross impaired assets: "36.1", "36.9", "32.5", "30.0", "41.8", "40.8", "42.9", "32.6".
  - Total provision to gross impaired assets: "110.4", "198.1", "105.6", "137.9", "131.8", "250.8", "125.6", "167.8".
  - Cash to total assets: "5.3", "5.3", "2.7", "2.8", "2.0", "1.6", "0.7", "0.9".
  - Cash and due from banks to total assets: "6.4", "7.4", "7.2", "9.9", "4.4", "3.0", "4.8", "6.3".
- The aggregate Tier 1 capital of the smaller domestic banks was more than "10 percent" as of March 2009.

Asset-quality trends and segmentation
- Past due loans plus impaired assets rose to around "1 percent" of total assets for the four largest banks as of March 2009.
- For all banks, past due plus impaired assets reached almost "1½ percent" of assets in March 2009.
- For smaller domestic banks, past due plus impaired assets reached nearly "3 percent".
- Causes for smaller banks’ deterioration: relatively large exposures to some lower quality commercial property and higher share of low-doc lending.
- Note: As of September 2008, total commercial property exposures of smaller domestic banks amounted to "$A 33 billion"—about "12 percent" of smaller banks’ total assets—with specific provisions and security held accounting for "97 percent" of impaired commercial property exposures.

Liquidity and funding structure
- Wholesale funding accounts for about "50 percent" of total funding, of which about "60 percent" is offshore.
- Financial institutions short-term external debt (residual maturity basis) estimated at about "$A 400 billion (35 percent of GDP)" in March 2009.
- Table 3 liabilities (percent of total) series:
  - Deposits: "41.5", "39.3", "42.5", "45.3" (Dec 06, Dec 07, Dec 08, Jun 09).
  - Domestic wholesale funding: "23.3", "26.7", "24.7", "21.1".
  - Offshore wholesale funding: "29.0", "28.2", "28.7", "30.0".
  - Securitization: "6.2", "5.8", "4.1", "3.6".
  - Equity (as a percent of total liabilities) memorandum item: "5.8", "5.6", "6.2", "6.6".
- Banks raised about "$A 140 billion" between December 2008 and early July 2009 after deposit and wholesale funding guarantees were established in October 2008.
- Banks have started to increase medium-term funding in response to liquidity and rollover risks.

### How would banks handle a jump in corporate defaults? (CCA analysis)
- Nonfinancial corporate sector condition:
  - Entered the crisis with moderate leverage and strong balance sheets.
  - Leverage (debt to assets and debt to equity ratios) remained stable and broadly similar to other advanced countries.
  - Profitability improved since the late 1990s; liquidity increased.
  - Market-based indicators show increased corporate solvency risks since 2008, but risks remain manageable.
- Methodology:
  - Contingent claim approach (CCA) used to estimate forward-looking risk indicators (distance to distress, probability of default) combining balance sheet data and share prices.
  - Translation of corporate expected losses into bank sector losses assumes equal exposure across banks (all banks suffer same relative losses).
- CCA findings and quantitative results:
  - Expected corporate default losses could amount to around "2 percent of GDP" based on historical recovery rates of around "40 percent", lower than other advanced countries in Asia (Figure 4).
  - Banks’ losses could amount to about "2 percent of total March 2009 loans" (Table 4), less than for other countries in the region (Figure 5).
- Caveats in interpreting CCA results:
  - Recent high equity market volatility could magnify distance to default and probability of default measures.
  - Bank lending to the nonfinancial corporate sector slowed significantly in 2009, suggesting that leverage has also declined since 2008.
  - Nonfinancial corporate sector raised significant equity capital during Q1 2009—private nonfinancial companies raised over "$A 18 billion" in shares and other equity in Q1 2009, almost twice the amount raised during Q1 2008.

Quantified CCA table excerpt (Table 4, Effects of Corporate Sector Distress on the Banking Sector: CCA Results)
- Columns and interpretation from source text:
  - Losses 1 Year Ahead, Bank Loans to Corporate Sector, Loan-Loss Provisions, Implied Additional NPL/Losses in Percent of Total Bank Loans, and related calculations are reported.
- Selected figures from Table 4 (Australia; columns reported as of March 2009 in source):
  - Bank Loans to Corporate Sector: "5.9" (in billions of Australian dollars in source presentation).
  - Loan-Loss Provisions: "1,651" (in billions of Australian dollars in source presentation).
  - Implied Additional NPL/Losses of Total Bank Loans: "1.7", "29.1", "1.8" as presented in table formula rows (see source for column mapping and formula E = A*(B-D)*(C/B) E/B).
- Notes on Table 4:
  - Sources: MKMV-Credit Edge data as of April 29, 2009; Fund staff calculations.
  - The CCA analysis covers listed companies only.
  - For New Zealand, columns B, C, and D report data as of September 2008; for Australia, B, C, and D report data as of March 2009.

### International comparison and contextual findings
- Australian large banks are less leveraged than banks in comparable countries (Section IV reference).
- Expected losses from corporate sector distress in Australia are lower relative to many comparator economies in advanced Asia and ASEAN4 (Figures 4 and 5 in source).
- Confidence-supporting policy measures (deposit and wholesale guarantees) facilitated substantial bond issuance and rollover of short-term debt.

_italic_Source: _wp09223 - References_

### 14.      Australian banks are compared in this section with international banks using

### _wp09223 - 14.      Australian banks are compared in this section with international banks using

### International comparison: leverage, deposit, and liquidity ratios
- Analysis uses mostly 2008 Bankscope measures for 60 large banks to compare Australian banks internationally (leverage, deposit, liquidity).
- Findings:
  - Australian banks have stronger leverage positions than major Canadian banks (Table 5).
  - Australian banks have weaker deposit and liquidity ratios than major Canadian banks (Tables 6 and 7).
  - Compared to the median of large international banks:
    - The four Australian banks are in the upper half of the sample in terms of leverage.
    - The four Australian banks are around the median in terms of deposit and liquidity ratio.
  - Conclusion: Australian banks seem to be among the stronger institutions, roughly in the same place as their Canadian counterparts.

- Selected exact figures from Tables (medians and ranks preserved as presented):
  - Equity/Total Assets (Median of banks in the sample): examples for Australian banks in 2008 — National Australia Bank 5.7; Australia and New Zealand Banking Group 5.5; Commonwealth Bank of Australia 5.1; Westpac Banking Corporation 4.5. (Table 5)
  - Deposits/Total Assets (Median of banks in the sample): examples for Australian banks in 2008 — Commonwealth Bank of Australia 57.7; National Australia Bank 55.3; Australia and New Zealand Banking Group 46.5; Westpac Banking Corporation 43.1. (Table 6)
  - Liquid Assets/Total (Median of banks in the sample): examples for Australian banks in 2008 — Westpac Banking Corporation 10.0; Commonwealth Bank of Australia 6.2; National Australia Bank 4.6; Australia and New Zealand Banking Group 3.4. (Table 7)

### Market-based indicators and sector co-movement
- Market indicators show strong co-movement among the four major Australian banks.
- Australian and Canadian banks show strong correlation in equity indicators and divergence from other advanced-country groups.
- Charts referenced:
  - Bank Equity Prices (March 2007=100) — strong co-movement among ANZ, Commonwealth Bank, Westpac, National Australia Bank.
  - Comparative equity price series for largest banks by jurisdiction (Canada, Australia, Euro Area, United States, United Kingdom).
  - Credit Default Swaps (5-year, average largest banks) — charts showing CDS spreads for Australian banks and cross-country comparisons.

### Limits of simple ratios for predicting equity performance
- Caution: liquidity, deposit, and leverage ratios have weak linear correlation with banks’ stock price evolution during the turmoil (Figures 7–9).
  - Figure 7 (Share Price Change (end-2006 - June 2009) vs Leverage (end-2006)): R^2 = 8E-06, regression y = 0.0003x + 5.6805.
  - Figure 8 (Share Price Change vs Deposit ratio (end-2006)): R^2 = 0.0058, regression y = 0.0338x + 56.138.
  - Figure 9 (Share Price Change vs Liquidity ratio (end-2006)): R^2 = 0.0187, regression y = 0.0605x + 18.347.
- Implication: assessments should pay attention to asset quality and additional complex measures (asset quality, supervision and regulation quality, market structure including securitization, competition).

### Banking sector financial soundness indicators (2007) — cross-country snapshot (selected figures)
- Assets to Tier 1 capital multiple examples: Sample Average 33.2; Australia 28.8; Canada 26.4; Iceland 185.1.
- Assets to total capital multiple examples: Sample Average 23.2; Australia 19.9; Canada 21.7; Iceland 156.9.
- Impaired loans to total loans examples: Australia 0.3; Canada 0.4; Iceland 6.2.
- Provisions to impaired loans examples: Australia 216.6; Canada 82.0; Iceland 156.7.
- Return on average assets examples: Australia 1.0; Canada 0.8; Iceland 2.1.
- Return on average equity examples: Australia 17.4; Sample Average 16.1.
- Liquid assets to deposits and ST funding examples: Australia 4.1; Canada 15.1; Sample Average 12.0.
- Notes: Table 8 presents these indicators; assets include off-balance sheet items; some figures expressed as multiples, not percent.

### Asset quality shock analysis — methodology
- Data source: Pillar 3 reports under the Basel II framework for the four large Australian banks (Australia and New Zealand Bank, Commonwealth Bank of Australia, National Australia Bank, Westpac).
- Loan portfolios consolidated into seven risk categories (I–VII).
- PD (probability of default) and LGD (loss given default) used to calculate likely one-year losses by risk category.
- Shock scenarios modeled:
  - Risk-shifting shock:
    - Corporate loans: shift risk categories up by one category (category I takes on PD/LGD of category II, etc.).
    - Mortgages: shift risk categories up by one and double the LGD floor to 40 percent.
    - Other loans: shift risk categories up by one; LGD floors not modified.
  - Six-times PD shock:
    - No risk-category shifting, but PDs increased six-fold; mortgages LGD floor of 40 percent.

### Results — risk-shifting and 40 percent mortgage LGD floor (Table 9 and Table 10)
- Expected total losses by category (Table 9, totals across categories): 181; 833; 2,770; 12,451; 9,506; 11,343; 5,348 (millions of Australian dollars) — total losses by category breakdown preserved as presented.
- Aggregate impact (Table 10):
  - Total losses (millions of Australian dollars): 42,432
  - Mortgage losses (millions of Australia dollars): 15,621
  - Corporate losses (millions of Australian dollars): 18,167
  - Provisions (millions of Australian dollars): 15,942
  - Total losses to capital (millions of Australian dollars): 26,490
  - Risk-weighted assets (millions of Australian dollars): 1,152,573
  - Loss as percent of risk-weighted assets: 2.3
  - Implied new total capital adequacy ratio (average of four banks): 9.2
  - Implied minimum new total capital adequacy ratio among the four banks: 8.3
  - Implied new tier 1 capital adequacy ratio (average of four banks): 6.1
  - Implied new tangible common equity to tangible asset ratio: 3.2
- Interpretation:
  - Banks appear resilient to this shock.
  - Even the hardest hit bank’s total capital adequacy ratio remains above the regulatory 8 percent minimum after this shock.
  - Mortgage insurance effects are not included and could reduce mortgage default impacts.

### Results — six-times PD increase and 40 percent mortgage LGD floor (Tables 11 and 12)
- Expected total losses by category (Table 11, totals across categories): 287; 2,091; 3,385; 21,529; 15,451; 11,261; 5,348 (millions of Australian dollars).
- Aggregate impact (Table 12):
  - Total losses (millions of Australian dollars): 59,353
  - Mortgage losses (millions of Australia dollars): 21,796
  - Corporate losses (millions of Australian dollars): 25,855
  - Provisions (millions of Australian dollars): 15,942
  - Total losses to capital (millions of Australian dollars): 43,411
  - Loss as percent of risk-weighted assets: 3.8
  - Implied new total capital adequacy ratio (average of four banks): 7.8
  - Implied minimum new total capital adequacy ratio among the four banks: 6.6
  - Implied new tier 1 capital adequacy ratio (average of four banks): 4.6
  - Implied new tangible common equity to tangible asset ratio: 2.5
- Interpretation:
  - A six-fold increase in PDs would be needed to reduce the total average capital adequacy ratio below 8 percent.
  - In this scenario average Tier 1 capital would remain above the regulatory minimum of 4 percent, and TCE would fall to 2½ percent.
  - Two banks’ total capital adequacy ratios would shrink below the 8 percent regulatory minimum.

### Comparison with contingent claim analysis (CCA)
- Corporate loan losses (Table 13) — before provisioning for losses:
  - Risk-shifting and LGD floor shock (Table III.10): 18.2 (In billions of Australian dollars), 1.1 (Percent of Banking Sector Loans)
  - Six-times PD increase and LGD floor shock (Table III.12): 25.9, 1.6
  - Contingent claim based analysis (Table III.4): 29.1, 1.8
- Interpretation: The CCA based estimate of corporate losses is close to the corporate loan loss impact of the more severe shock (six-times PD increase).

*Source: IMF staff analysis and tables from the provided chapter content.*

### 25.      Though banks seem resilient, more complex stress testing is needed. The above

### 25.      Though banks seem resilient, more complex stress testing is needed. The above

### Summary of assessment
- The shocks described do not constitute a rigorous stress test and the results are only indicative of the health of the banking sector.
- APRA has regularly stress tested the banking sector but it would be advisable to undertake more extreme scenarios than applied in the past and to include Australia’s overseas subsidiaries.
- In particular, stress tests should include a more protracted and serious macroeconomic downturn than what was applied in the 2006 Financial Stability Assessment Program.

### Contingent Claim Analysis (CCA) framework overview
- The CCA is a risk-adjusted balance sheet framework where equity and risky debt of a firm or financial institution derive their value from assets.
- In this framework, first proposed by Robert Merton (1973) and by Black and Scholes (1973), the total value of assets is equal to the market value of equity and risky debt.
- Asset values are uncertain and in the future may decline below the point where debt payments on scheduled dates cannot be made. Debt is “risky” since there is a chance of default.
- The assets are stochastic and evolve according to a “distress barrier”.
- See Gray and Malone (2008) for a comprehensive analysis of the CCA framework.

### Black-Scholes-Merton (BSM) application and measures
- The BSM model is used to estimate the default probability and distance-to-default.
- BSM derive the market’s assessment of default risk for a company from its equity price, assuming that the market price reflects investors’ correct calculation of default risk.
- The BSM default probabilities show the theoretical probability of default one year-ahead.
- Distance-to-default shows how much the asset value needs to fall one-year-ahead for a firm to default given its current balance sheet position; it is reported in terms of the number of standard deviations of asset returns: the higher this number, the lower the BSM probability of default.
- According to the BSM model, the logarithm of a firm’s assets is assumed to follow the standard Brownian motion.

### Distance-to-default (DtD) formula and computation notes
- The distance to default within one year is equal to (DtD)= 
A
A
BA
σ
σ
μ










2
)log()log(
3
2
, 
where:
  - A is total assets,
  - B is the default barrier measured as short-term debt plus one half of long-term debt plus interest payments,
  - μ is the expected return on assets (based on last year's annual capital gain including dividends),
  - σA is the standard deviation of the asset return.
- Because DtD is normally distributed with mean zero, we add 3 to the calculated DtD measure so that the reported DtD is always positive.
- DtD is calculated from pooled data, adding all inputs into a synthetic company at the country level.
- Asset values and the standard deviation of asset returns are derived using the Black-Scholes-Merton option pricing formula, with stock prices and their volatility as inputs.

### Computation of banks’ expected losses from corporate sector distress
- Banks’ expected losses from corporate sector distress (Figures 4 and 5) were calculated using information from Moody’s KMV implied CDS (EICDS) spreads and banks’ exposure to the corporate sector.

Calculation steps:
- Expected losses for the corporate sector one year ahead embedded in EICDS spreads were calculated using the contingent claim analysis framework.
- The corporate sector expected losses were expressed as ratios of the corporate sector’s total liabilities. It was then assumed that all the corporate sector’s creditors would suffer the same relative losses in their books in order to overcome lack of more precise calculation on the seniority structure of the debt and on the relative importance of domestic versus foreign financing sources.
- Banks’ current performing loans to the corporate sector were calculated. In the absence of information on banks’ current provisions for losses on loans to the corporate sector, banks’ overall provisions for losses were subtracted from the current stock of their loans to the corporate sector, and the resulting amount was scaled by banks’ exposure to the corporate sector.
- The relative losses calculated in the second step were multiplied by the current stock of performing loans to the corporate sector. The resulting amount was the expected increase in banks losses stemming from banks’ exposure to the corporate sector.

### Key implications and recommendations
- More extreme and protracted macroeconomic downturn scenarios should be included in stress testing exercises.
- Stress testing should explicitly incorporate Australia’s overseas subsidiaries.
- Use of CCA/BSM measures (one-year-ahead default probabilities and distance-to-default) and incorporation of market-implied measures (EICDS spreads) can quantify expected corporate-sector losses and translate them into bank-level expected losses given exposure profiles.

*Source: _wp09223 - 25.      Though banks seem resilient, more complex stress testing is needed.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09223.pdf_