OCC Bulletin 2012-33| October 18, 2012
Community Bank Stress Testing: Supervisory Guidance
Chief Executive Officers of All National Banks, Federal Savings Associations, Department and Division Heads, Examining Personnel, and Other Interested Parties
See OCC Bulletin 2021-21, which explains that the stress test tool for income-producing CRE loan portfolios has been terminated and is no longer on the OCC's BankNet website.
The Office of the Comptroller of the Currency (OCC) is issuing Bulletin OCC 2012-33, "Community Bank Stress Testing: Supervisory Guidance," to provide guidance to national banks and federal savings associations (collectively, banks) with $10 billion or less in total assets on using stress testing to identify and quantify risk in loan portfolios and help establish effective strategic and capital planning processes.
Community banks, regardless of size, should have the capacity to analyze the potential impact of adverse outcomes on their financial conditions. The OCC's guidance describes various types of stress test methods that community banks may use and provides one example of a simple stress test framework to consider. The OCC encourages community banks to adopt a stress test method that fits their unique business strategy, size, products, sophistication, and overall risk profile.
Bank management and examiners should use this guidance in conjunction with the "Concentrations of Credit" booklet in the OCC's Comptroller's Handbook series, OCC Bulletin 2012-16, "Guidance for Evaluating Capital Planning and Adequacy," and the Interagency Final Guidance on "Concentrations in Commercial Real Estate Lending, Sound Risk Management Practices." These documents provide additional information on stress testing objectives and methods.
Background and Supervisory Expectations
The recent financial crisis demonstrated how unexpected economic downturns and rapid deterioration in market conditions can significantly harm a bank's financial condition and economic viability. Concentrations of credit, particularly in commercial real estate (CRE) loans for acquisition, construction, and land development purposes, have been a common factor in bank failures during stressful periods, especially for community banks. Other factors that have played a role in weakened bank conditions and failures include inadequate capital, dependence on brokered deposits, and dependence on assets whose valuations are highly sensitive to volatility in energy and commodity prices, interest rates, or farmland prices.
Sound risk management practices should include an understanding of the key vulnerabilities facing banks. For several years, supervisors have used the term "stress testing" in guidance and handbooks to refer to and encourage banks to incorporate this practice.1 Well-managed community banks routinely conduct interest rate risk sensitivity analysis to understand and manage the risk from changes in interest rates. Many community banks, however, do not have similar processes in place to quantify risk in loan portfolios, which often are the largest, riskiest, and highest earning assets.
The Dodd–Frank Wall Street Reform and Consumer Protection Act of 2010 (Dodd–Frank) requires annual stress testing for banks with assets greater than $10 billion, and the federal banking agencies have issued separate guidance for stress testing standards in those institutions.2 The OCC recognizes that while some large community banks may find that guidance useful, community banks are not required to use such holistic stress testing.3
The OCC, however, does consider some form of stress testing or sensitivity analysis of loan portfolios on at least an annual basis to be a key part of sound risk management for community banks. Community banks that have incorporated such concepts and analyses into their credit risk management and strategic and capital planning processes have demonstrated the ability to minimize the impact of negative market developments more effectively than those that did not use stress testing.
Community bank management can use stress testing to establish and support reasonable risk appetite and tolerances, set concentration limits, adjust strategies, and appropriately plan for and maintain adequate capital levels. Bank management should mitigate identified risks and vulnerabilities through such actions as increased portfolio monitoring, adjusted underwriting standards, selling or hedging assets, and increasing capital. In addition, bank management should use the results of stress tests to establish appropriate action plans that address risks when the results are inconsistent with risk tolerance levels and the bank's overall strategic and capital plans.
Stress Testing Methods
Stress testing is not a new concept for community banks. Scenario or sensitivity analysis and stress testing requirements have been used for subprime lending, interest rate risk, and liquidity management for several years.4 A community bank's approach to stress testing should fit its unique loan portfolio strategy, size, loan types, composition, operations, and management. Given the smaller scale and lesser complexity of most community banks, assessing portfolio risk and capital vulnerability can be relatively simple.
The OCC does not endorse a particular stress testing method for community banks. Stress tests do not need to involve sophisticated analysis or third-party consultative support. Effective methods can range from a single spreadsheet analysis to a more sophisticated model, depending on portfolio risk and the complexity of the bank. Community banks may need to make only modest enhancements to existing risk management practices and techniques to ensure that potential adverse outcomes are appropriately considered.
Refer to appendix A of this bulletin for additional information on a variety of stress testing methods that community banks may consider.
Regardless of the testing method used, an effective stress test has common elements that a community bank should consider. These include
- asking plausible "what if" questions about key vulnerabilities;
- making a reasonable determination of how much impact the stress event or factor might have on earnings and capital; and
- incorporating the resulting analysis into the bank's overall risk management process, asset/liability strategies, and strategic and capital planning processes.
For most community banks, a simple, stressed loss-rate analysis based on call report categories may provide an acceptable foundation to determine if additional analysis is necessary. Banks should primarily focus on concentrations of credit or loan portfolio segments that are significant to the overall business strategy. A community bank may be able to examine on a single spreadsheet its key assets and liabilities and link these to income and funding requirements. The loss stress rates used may be derived from a review of historical loss experience during previous stressful periods, historical market experience, or other estimates.5 Appendix B contains an example of this type of stress test method.
The potential adverse impact on earnings, loan loss reserves, and capital revealed by the stress test should be evaluated to understand the impact on capital and to ensure the bank can maintain appropriate capital commensurate with its overall risk profile. If the stress test reveals critical vulnerabilities, management and the board should take steps to mitigate those risks through such means as modifying loan growth, revising the risk tolerance strategy, adjusting the portfolio mix and underwriting criteria, altering concentration limits or other policies and procedures, and strengthening capital.
Community banks can conduct stress tests on identified credit concentrations,6 on other loan portfolio segmentations, and at the individual loan level. For example, a review by senior management may reveal two or three key concentrations in the loan portfolio, such as loans dependent on a type of agribusiness, loans with construction-related risks, long-term fixed rate municipal securities, commercial mortgage loans dependent on local market values, or consumer residential loans. Selecting the appropriate factors to stress depends on the nature of the bank's concentration risk. Banks with an ability to perform migration analysis of credit risk ratings may be able to use that knowledge in evaluating changes in credit quality across different loan portfolios, because higher grade credits typically withstand market stresses better than lower grade credits.
While problem credits are routinely the focus of risk management, the potential effects on credit concentrations or other portfolio segments are the focus of portfolio level stress testing. Insight gained from such broad portfolio analysis can provide bank management useful information for strategic and capital planning initiatives and a better understanding of capital at risk. Furthermore, the appropriate time frame for a stress test scenario should be at least a two-year projection because, in any given credit cycle, losses generally emerge over a two-year period following the downturn. For example, in the recession that lasted from December 2007 to June 2009, the economy experienced a run of more than six quarters of weak or negative gross domestic product before commercial credit quality indicators reached their worse performance, and loan charge-off rates did not return to more normal historical rates until nine or 10 quarters after the initial economic downturn.
In some cases, a bottom-up, loan-by-loan analysis may help identify particular concerns. For more complex portfolios, such as commercial mortgages or construction loans, further segmentation may be helpful to differentiate various levels of risk. For example, management may link commercial mortgages to debt service coverage and loan-to-value ratios to project potential loss under possible adverse circumstances. Construction loans may be sensitive to particular variables such as selling rates, leasing activity, or oversupply.
Per OCC Bulletin 2006-46, "Interagency Guidance on CRE Concentration Risk Management," banks that exceed certain CRE concentration thresholds are expected to use more robust stress testing practices to effectively manage the concentrations and maintain adequate capital. The OCC's new stress test tool for income-producing CRE loan portfolios is available on the OCC's BankNet website along with other tools banks may use to stress test other types of loans. Appendix C contains an example of factors to consider when conducting CRE loan stress tests or sensitivity analyses.
Stress Testing and Capital Planning
The OCC expects every bank, regardless of size or risk profile, to have an effective internal process to (1) assess its capital adequacy in relation to its overall risks, and (2) to plan for maintaining appropriate capital levels.7 Stress testing can be a prudent way for a community bank to identify its key vulnerabilities to market forces and assess how to effectively manage those risks should they emerge.
If the results of a stress test indicate that capital ratios could fall below the level needed to adequately support the bank's overall risk profile, the bank's board and management should take appropriate steps to protect the bank from such an occurrence. This may include establishing a plan that requires closer monitoring of market information, adjusting strategic and capital plans to mitigate risk, changing risk appetite and risk tolerance levels, limiting or stopping loan growth or adjusting the portfolio mix, adjusting underwriting standards, raising more capital, and selling or hedging loans to reduce the potential impact from such stress events.
John C. Lyons Jr.
Senior Deputy Comptroller and Chief National Bank Examiner
Stress Testing Methods and Approaches
In broad terms, stress testing can refer to many different types of methods and applications, including transaction stress testing, portfolio stress testing, enterprise stress testing, and reverse stress testing. A bank can use a variety of stress test methods to evaluate loan portfolio risk and the potential impact on earnings and capital based on its unique risk profile.
Many of these types of stress testing fall under the general category of scenario analysis.8 Scenario analysis refers to stress testing where a bank applies historical or hypothetical scenarios to assess the impact of various events and circumstances, including extreme ones. The variables are usually determined by the scenarios, which involve some kind of coherent, logical narrative or story portraying why certain events and circumstances can occur and in which combination and order they may occur. Examples of common scenarios include a severe recession, loss of a major client, or a localized economic downturn.
Scenario analysis can be applied at various levels of the banking organization, such as the transaction level, portfolio level, and enterprise-wide. It can also be applied in "reverse" to develop the type of scenarios that could result in critical harm to the institution.
Transaction stress testing is a method that estimates potential losses at the loan level by assessing the impact of changing economic conditions on a borrower's ability to service debt. Transaction level scenario stress testing can help in a "bottom up" analysis to gauge a borrower's vulnerability to default and loss, foster early problem loan identification and strategic decision making, and strengthen strategic decisions about key loans.
Portfolio stress testing is a method that helps identify current and emerging risks and vulnerabilities within the loan portfolio by assessing the impact of changing economic9 conditions on borrower performance, identifying credit concentrations, measuring the resulting change in overall portfolio credit quality, and ultimately determining the potential financial impact on earnings and capital. In a "bottom up" approach, this consists of aggregating the results of individual transaction level stress tests given changes in key variables driven by economic forecasts under one or more scenarios. In a "top down" approach, this consists of applying estimated stress loss rates under one or more scenarios to pools of loans with common risk characteristics.
Regardless of the method used, the different scenarios should include a projected base case and at least one or more adverse scenario(s) based on macro and local economic data. A bank may have to develop different variable assumptions for pools of loans with similar characteristics, such as geography and collateral type, within each scenario. The process of stress testing portfolios can aid in strategic decision making, credit policy development, strengthen the quality of concentration risk management, support reserve methodology, and determine regulatory capital at risk.
Loan migration analysis can be used by banks with larger portfolios and more comprehensive internal databases to evaluate how a downward migration in internal loan ratings, consistent with migrations that might be expected during adverse economic conditions, would impact asset quality, earnings, and capital. This analysis would also assist banks in determining possible actions to address potential deterioration in their portfolios.
Enterprise-level stress testing is a method that considers multiple types of risk and their interrelated effects on the overall financial impact under a given economic scenario. These risks include, but are not limited to, credit risk within loan and security portfolios, counter-party credit risk, interest rate risk, and changes in the bank's liquidity position. In its most basic form, enterprise-level stress testing can be performed by aggregating portfolio stress testing results, while considering the related impacts from interest rate risk and liquidity stress tests; however, the assumptions used in each type of stress test should be consistent and derived from the same economic scenarios. As with portfolio stress testing, enterprise stress testing should consider a projected base case and at least one adverse scenario. The sophistication of the enterprise-level stress testing should be commensurate with the size and complexity of the bank.
Reverse stress testing is a method under which the bank assumes a specific adverse outcome, such as suffering credit losses sufficient to cause a breach in regulatory capital ratios, and then deduces the types of events that could lead to such an outcome. This type of analysis (e.g. a "break the bank" scenario) can help a bank consider scenarios beyond normal business expectations and challenge common assumptions about performance and risk mitigation strategies. For example, if the bank has a highly concentrated yet geographically diverse construction portfolio, a reverse stress test may help a bank identify conditions (changes in key variables) that would cause losses across each geographic segment sufficient to cause capital ratios to fall below regulatory minimum levels. Having identified such scenarios, bank management can consider how likely those conditions are, make contingency plans, or take other steps to mitigate the identified risks.
Constructing a Basic Portfolio Level Stress Test
The following example using stress loss rates in testing is for illustrative purposes only.
For many community banks the following methodology can serve as a starting point for stress testing analysis. Other methods that quantify potential risk, however, are also acceptable. Community banks should customize this example for their own loan portfolios and may consider supplementing this analysis or use other stress test alternatives based on their unique portfolio characteristics. The example has three sections.
Section 1 Estimated Loan Portfolio Stress Losses
Objective: This section estimates the potential loan losses over a two-year stress test horizon for the entire loan portfolio. There are four components in this section.
Loan Portfolio Categories
This component stratifies the loan portfolio into segments with similar loss characteristics. Management should consider the bank's material exposures and loss correlations across the entire credit portfolio with particular emphasis on key vulnerabilities, such as concentration risk. Management may approach this segmentation using its portfolio management information systems or through its call report categories on Schedule RC-C. In this example, loans secured by real estate are stratified for more detail, but smaller call report categories are grouped into the "All Other Loans" category. Banks with material exposures in other asset categories that may decline significantly in value, such as the investment portfolio, may also want to consider those assets in the stress test.
Quarter-End Loan Portfolio Balances
This component aggregates the quarter end loan balances by loan portfolio categories selected above. A bank can run a stress test at any time, but management information systems and reporting at community banks are often more robust at quarter end.
Stress Period Loss Rates
This component applies an aggregate loss rate over the stress test horizon to the loan portfolio segments. Typical economic downturns result in credit cycle impacts that evolve over a two-year or longer period, so the analysis should consider at least two years. Management should assess what might happen to loss rates for the loan portfolio with particular emphasis on material exposures in an economic downturn. The bank's own historical experience through previous recessions or financial stress periods may provide a starting point to determine stressed loss rates, or the bank may refer to outside references for ranges of performance for community banks. The severity of the loss rates should consider economic conditions. For example, history has shown that loan portfolio risk may be increasing despite extended periods of minimal or below average loss rates. This may mask the possibility that future losses could exceed a bank's historical losses.
In this example, we selected loss rates within the ranges provided in the OCC provided historical loan loss benchmark data. Banks would consider the current economic environment, underwriting standards, and recent collateral appreciation in addition to other unique bank specific factors when selecting a loss rate. Bankers can use their own analysis or the OCC provided historical loan loss data as a reference for loss rates.
Stress Period Losses
This component represents the multiplication of the loan balances and the loss rates. The next section uses the sum of these losses.
Section 2 Estimated Impact on Earnings
Objective: This section estimates the potential impact to net income from the stress scenario over the two-year period. There are five components in this section.
Pre-provision Net Income
This component aggregates all revenue and expense implications except loan loss provisions and income taxes over the two-year stress horizon. Management should consider what might happen to revenue and expense levels in a stressed environment. This would include the impact of higher nonaccruals and increased collection costs. Management should also consider the potential for funding issues on the liability side of the balance sheet. Management should use information from their interest rate risk and liquidity stress tests to assist in estimating the potential decline in revenue and net interest income during the two-year stress period.
Provision Expense to Cover Stress Losses
This component represents the estimated stress losses from section 1 and does not consider the additional provision necessary to increase the Allowance for Loan and Lease Losses (ALLL) to an adequate level.
Provision to Maintain an Adequate ALLL
This component is an estimate of the additional provision expense necessary to maintain the ALLL at an adequate level at the end of the stress period. In stressed environments the risk in the loan portfolio increases significantly causing a need for a higher ALLL. Management can use its experience during recessionary periods to estimate an appropriate ending ALLL. The incremental amount of provision expense necessary to maintain the ALLL at an adequate level is entered on this line.
Income Tax Expense (Benefit)
This component estimates income tax adjustments applicable to pre-tax income resulting from the stress scenario. This amount is estimated using the bank's effective tax rate.
This component represents the aggregate net income over the two-year stress test period. The next section uses this number.
Section 3 Estimated Impact of Stress on Capital
Objective: This section estimates the hypothetical impact on capital of the stressed environment. The example uses Tier 1 capital and the Tier 1 leverage ratios to help analyze the potential change in capital caused by a stress scenario. Banks can also review the changes in other relevant capital measures, such as the potential change in the common equity ratio, to assess the results of the stress test. This section has five components.
Tier 1 Capital
This component represents the quarter end amount of Tier 1 capital at the start of the stress test.
Net Change in Tier 1 Capital
This component applies only to the stress scenario and represents the reduction in capital generated by stress scenario. This amount is the net income calculated in the stress scenario from section 2.
Adjusted Tier 1 Capital
This component represents the sum of the previous components in the section and is the bank's hypothetical capital after the stress period. Management uses this number to estimate the Tier 1 leverage ratio below.
Quarterly Average Assets
This component represents the estimate of the bank's quarterly average assets at the end of the stress scenario. Management can use historical information to estimate this component. This example anticipates the bank maintains existing relationships but generates no growth during the stress period. In addition, no asset sales are projected during the stress period. Therefore, in the example, the average assets decline is in line with reductions in the loan portfolio from the stress losses.
Tier 1 Leverage Ratio
This component represents a measure of the bank's capital levels at the start and end of the stress scenario. Bank management should analyze this change in relation to the bank's concentration management, capital planning, and strategic planning processes. Management can also use this analysis to monitor risk tolerances and test the impact of business strategies to increase/decrease exposures or expand into new products. There is significant value in looking at the marginal changes in risk levels indicated by stress testing as management consistently conducts stress testing over time.
|1. Estimated Loan Portfolio Stress Losses|
|Loan Portfolios from Call Report
|Quarter End As of Date
|Two-Year Stress Period Loss Rate %||Two-Year Stress Period $ Losses|
|Loans secured by type of real estate|
|a. Construction and development||100||20%||20|
|c. 1-4 family housing||100||4%||4|
|d. Multifamily housing||75||16%||12|
|e. Nonfarm nonresidential property||100||8%||8|
|Agricultural production and farmer loans||40||6%||2.4|
|Commercial and industrial||60||14%||8.4|
|All other loans||25||4%||1|
|2. Estimated Impact of Stress on Earnings|
|Previous Two Years Actual||Pro Forma Stress Period|
|Pre-provision net income||34.5||30|
|Less provision to cover two-year losses||12||61.8|
|Less provision to maintain adequate ALLL||0||10|
|Income tax expense (benefit)||5.5||(14.6)|
|3. Estimated Impact of Stress on Capital|
|Quarter End As of Date||Pro Forma Stress Period|
|Tier 1 Capital $||88||88|
|Net change in Tier 1 Capital from stress period (net income from Step 2)||N/A||(27.2)|
|Adjusted Tier 1 Capital $||88||60.8|
|Quarterly Average Assets $||800||738|
|Tier 1 Leverage Ratio %||11%||8.2%|
*Data used in this example are not real and are for illustrative purposes only.
Possible Real Estate Stress Testing Factors
The following table lists characteristics and variables common to particular property types that may be considered when evaluating the impact of a stress period on the property type. These characteristics may be part of a sub-schedule of stress testing a portfolio of CRE loans.
|STRESS FACTORS||Land Acquisition & Development||Residential Construction||Commercial Construction (1)||Multi-Family Residential||Residential & Commercial Condominium Construction||Leased Commercial Office Space||Leased Industrial and Warehouse Distribution||Retail Boxes & Strip Malls|
|Absorption / Sell-Out Rates||✓||✓||✓||✓|
|Utility and/or Energy Costs||✓||✓||✓||✓||✓||✓||✓||✓|
(1) Includes multi-family residential, leased commercial office space, leased industrial and warehouse distribution, and retail boxes and strip malls.
Source: Federal Reserve Bank of Philadelphia, "Stress Testing: A Risk Management Tool for Commercial Real Estate Loan Concentrations," Third Quarter 2008.
1 Refer to the OCC Comptroller's Handbook, "Loan Portfolio Management" booklet, April 1998, pages 30–32.
2 Dodd–Frank requires all banking organizations with total consolidated assets of more than $10 billion to conduct annual stress tests in accordance with regulations issued by their primary federal financial regulatory agencies. On January 24, 2012, the OCC issued a notice of proposed rulemaking to implement this requirement (77 FR 3408). Separately, on May 14, 2012, the federal banking agencies issued stress testing guidance for banking organizations with more than $10 billion in total consolidated assets (77 FR 29458). See OCC Bulletin 2012-14, "Interagency Stress Testing Guidance."
3 Refer to OCC News Release 2012-76, "Statement to Clarify Supervisory Expectations for Stress Testing by Community Banks."
5 Refer to the "Banking Tools" section of BankNet, which is available to all national banks and federal savings associations. Benchmark historical stress loan performance ranges are provided as a reference that bankers can use to supplement bank specific information, especially for adverse scenarios.
6 OCC's Comptroller's Handbook, "Concentrations of Credit" booklet, December 2011.