Capital Planning At Bank Holding Companies

• The Federal Reserve has previously noted the importance of capital planning at large and   complex bank holding companies (BHCs).

• Capital is central to a bank holding companies’ ability to absorb unexpected losses and   continue to lend to creditworthy businesses and consumers. It serves as the first line of defense   against losses, protecting the deposit insurance fund and taxpayers.

• As such, a large BHC’s processes for managing and allocating its capital resources are critical   not only to its individual health and performance, but also to the stability and effective   functioning of the US financial system.

Capital Plan Rule

• The Federal Reserve’s Capital Plan Rule and the associated annual Comprehensive Capital   Analysis and Review (𝐶𝐶𝐴𝑅) have emphasized the importance the Federal Reserve places on   bank holding companies’ (BHC) internal capital planning processes, and on the supervisory   assessment of all aspects of these processes, which is a key element of a supervisory program   that is focused on promoting resiliency at the largest bank holding companies.

• These initiatives have focused not just on the amount of capital that a BHC has, but also on   the internal practices and policies a firm uses to determine the amount and composition of   capital that would be adequate, given the firm’s risk exposures and corporate strategies as well   as supervisory expectations and regulatory standards.

• The Federal Reserve’s Capital Plan Rule requires all US – domiciled, top-tier BHCs with total   consolidated assets of $50 billion or more to develop and maintain a capital plan supported by   a robust process for assessing their capital adequacy.

• The Federal Reserve’s assessment of a BHC’s capital planning process includes an evaluation   of the risk-identification, -measurement, and -management practices that support the BHC’s capital planning and stress scenario analysis, an assessment of  stressed loss and revenue estimation practices, and a review of the governance and controls around these practices.
• The preamble to the Capital Plan Rule outlines the elements on which the Federal Reserve   evaluates the robustness of a BHC’s internal capital planning – also referred to as the capital   adequacy process, or “𝐶𝐴𝑃”.

Seven Principles Of An Effective Capital Adequacy Process

• Principle 1: Sound foundational risk management

The BHC has a sound risk-measurement and risk-management infrastructure that supports the identification, measurement, assessment, and control of all material risks arising from its exposures and business activities.

• Principle 2: Effective loss-estimation methodologies

The BHC has effective processes for translating risk measures into estimates of potential losses over a range of  stressful scenarios and environments and for aggregating those estimated losses across the BHC.

• Principle 3: Solid resource-estimation methodologies

The BHC has a clear definition of available capital resources and an effective process for estimating available capital resources (including any projected revenues) over the same range of stressful scenarios and environments used for estimating losses.

• Principle 4: Sufficient capital adequacy impact assessment

The BHC has processes for bringing together estimates of losses and capital resources to assess the combined impact on capital adequacy in relation to the BHC’s stated goals for the level and composition of capital.

• Principle 5: Comprehensive capital policy and capital planning

The BHC has a comprehensive capital policy and robust capital planning practices for establishing capital goals, determining appropriate capital levels and composition of capital, making decisions about capital actions, and maintaining capital contingency plans.

• Principle 6: Robust internal controls

The BHC has robust internal controls governing capital adequacy process components, including policies and procedures; change control; model validation and independent review; comprehensive documentation; and review by internal audit.

• Principle 7: Effective governance

The BHC has effective board and senior management oversight of the 𝐶𝐴𝑃, including –

i. Periodic  review  of  the  BHC’s  risk  infrastructure  and  loss  –  and  resource-estimation methodologies

ii. Evaluation of capital goals

iii. Assessment of the appropriateness of stressful scenarios considered

iv. Regular  review  of  any  limitations  and  uncertainties  in  all  aspects  of  the  𝐶𝐴𝑃;  and approval of capital decisions.

Best Practices For Risk Identification

• BHCs should have risk-identification processes that ensure that all risks are appropriately accounted for when assessing capital needs.

• These processes should evaluate the full set of potential exposures stemming from both on-   balance sheet and off-balance sheet positions, including those that could arise from provisions   of  noncontractual support to off-balance-sheet entities, and risks conditional on changing   economic and financial market conditions during periods of stress.

• BHCs should have a systematic and repeatable process to identify all risks and consider the   potential impact to capital from these risks. In addition, BHCs should closely assess any   assumptions about risk reduction resulting from risk transfer and/or mitigation techniques, including, for example, analysis of the enforceability and effectiveness of any guarantees or netting.

• Stronger risk-identification practices include standardized processes through which senior   management regularly update risk assessments and review risk exposures and consider how   their risk exposures might evolve under a variety of stressful situations. For example, many BHCs maintain a comprehensive inventory of risks to which they are exposed and refresh it as conditions warrant (such as changes in the business mix and the operating environment) with input from various units across the BHC.
• Senior representatives from major lines of business, corporate risk management, finance and   treasury, and other business and risk functions with perspectives on BHC-wide positions and   risks provide input to the process. Consideration of the risks inherent in new products and   activities should be a key part of  risk-identification and risk-assessment programs, which   should also consider risks that may be associated with any change in the BHC’s strategic   direction.
• Risk measures should be able to capture changes in an institution’s risk profile – whether due to   a change in the BHC’s strategic direction, specific new products, increased volumes, changes in   concentration or portfolio quality, or the overall economic environment – on a timely basis.
• These risk measures should support BHCs’ assessments of  capital adequacy and may be   helpful in capital contingency plans as early warning indicators or contingency triggers, where appropriate.
• BHCs should be able to demonstrate how their identified risks are accounted for in their   capital planning processes. If  certain risks are omitted from the enterprise-wide scenario   analysis, BHCs should note how these risks are accounted for in other aspects of the capital   planning process.
• If  a BHC employs risk quantification methodologies in its capital planning that are not   scenario-based, it should identify which risks each of the methodologies covers, to facilitate   comparability and informed decision-making with respect to overall capital adequacy.
• BHCs with lagging practice do not transparently link their evaluation of capital adequacy to the   full range of  identified risks. These BHCs are not able to show how all their risks were   accounted for in their capital planning processes.
• In some cases, staff responsible for capital planning operate in silos and develop standalone   risk inventories not linked to the enterprise-wide risk inventory or to other risk governance   functions within their BHCs. This can lead to an underestimation of risks.

Best Practices Regarding Internal Control

• As with other aspects of key risk-management and finance area functions, a BHC should have   a strong internal control framework that helps govern its internal capital planning processes.   These controls should include –

1.Regular and comprehensive review by internal audit

2.Robust and independent model review and validation practices

3.Comprehensive documentation (including policies and procedures)

4.Change controls.

• A  BHC’s  internal  control  framework  should  address  its  entire  capital  planning  process, including –

1.The risk measurement and management systems used to produce input data

2.The models and other techniques used to generate loss and revenue estimates

3.The aggregation and reporting framework used to produce reports to management and boards

4.The process for making capital adequacy decisions.

• While some BHCs may naturally develop components of their internal capital planning along separate business lines, the control framework should ensure that BHC management reconciles the separate components in a coherent manner.
• The control framework should also help assure that all aspects of the capital planning process   are functioning as intended in support of robust assessments of capital needs. BHCs should   review the controls around capital planning on an integrated basis and apply them consistently.
• Management should respond quickly and effectively to issues identified by control areas and   devote appropriate resources to continually ensure that controls are functioning effectively.
• Internal audit should play a key role in evaluating internal capital planning and its various   components. Audit should perform a periodic review of  the full process, not just of  the   individual  components,  to  ensure  that  the  entire  end-to-end  process  is  functioning  in   accordance with supervisory expectations and with a BHC’s board of directors’ expectations as   detailed in approved policies and procedures.
• Internal audit should review the manner in which deficiencies are identified, tracked, and   remediated. Audit staff should have the appropriate competence and influence to identify and   escalate key issues, and the internal audit function should report regularly on the status of all aspects of  the capital planning process – including any identified deficiencies related to the BHC’s capital plan – to senior management and the board of directors.
• BHCs should provide a comprehensive and robust review of all components of the capital   planning process, including all of  the control elements noted earlier. BHCs with leading   internal audit practices around internal capital planning have strong issue identification and   remediation tracking as well. They also ensure that audit staff have strong technical expertise,   elevated stature in the organization, and proper independence from management.
• BHCs should conduct independent review and validation of all models used in internal capital   planning, consistent with existing supervisory guidance on model risk management. Validation   staff  should  have  the  necessary  technical  competencies,  sufficient  stature  within  the   organization, and appropriate independence from model developers and business areas, so that   they can provide a critical and unbiased evaluation of the models they review.
• The model review and validation process should include –

i. An evaluation of conceptual soundness

ii. Ongoing monitoring that includes verification of processes and benchmarking

iii.  An outcomes analysis

• BHCs should maintain an inventory of  all models used in the capital planning process,   including all input or “feeder” models that produce projections or estimates used by the models   that generate the final loss, revenue or expense projections.

• Consideration should be given to the validity of the use of a model under stressed conditions   as models designed for ongoing business activities may be inappropriate for estimating net   income  and  capital  under  stress  conditions.  BHCs  should  also  maintain  a  process  to   incorporate well-supported adjustments to model estimates when model weaknesses and   uncertainties are identified.

• BHCs continue to face challenges in conducting outcomes analysis of  their stress testing   models, given limited realized outcomes against which to assess loss, revenue, or expense   projections under stressful scenarios.

• BHCs should attempt to compensate for the challenges inherent in backtesting stress models   by conducting sensitivity analysis or by using benchmark or “challenger” models. BHCs should ensure that validation covers all models and assumptions used for capital planning purposes, including any adjustments management has made to the model estimates (management overlay).
• Supervisory reviews have found that, in general, BHCs should give more attention to model risk management, including strengthening practices around model review and validation.
• Nonetheless, some BHCs exhibit stronger practices in their capital planning, including –

1. Maintaining an updated inventory of all models used in the process

2. Ensuring that models had been validated for their intended use

3. Being transparent about the validation status of all models used for capital planning

4. Appropriately addressing any models that had not been validated (or those that had   identified weaknesses) by restricting their use or using benchmark or challenger models to   help assess the reasonableness of the primary model output.

• BHCs with lagging practices generally –

i. Are not able to identify all models used in the capital planning process

ii. Do not formally review all of the models or assumptions used for capital planning purposes (including some high-impact stress testing models)

iii.  Do not have validation staff that are independent and can critically evaluate the models.

• BHCs should ensure that they have policies and procedures covering the entire capital planning   process. Policies and procedures should ensure a consistent and repeatable process for all   components of the capital planning process and provide transparency to third parties regarding   this process.

• Policies should be reviewed and updated at least annually and more frequently when warranted.   There should also be evidence that management and staff  are adhering to policies and   procedures in practice, and there should be a formal process for any policy exceptions. Such   exceptions should be rare and approved by the appropriate level of management.

• BHCs should have internal controls that ensure the integrity of  reported results and the   documentation, review, and approval of all material changes to the capital planning process and   its components. A BHC should ensure that such controls exist at all levels of  the capital   planning process. Specific controls should be in place to –

i. Ensure that MIS are sufficiently robust to support capital analysis and decision-making, with sufficient flexibility to run ad hoc analysis as needed

ii. Provide for reconciliation and data integrity processes for all key reports

iii. Address the presentation of aggregate, enterprise-wide capital planning results, which   should describe any manual adjustments made in the aggregation process and how those   adjustments compensate for identified weaknesses

iv. Ensure  that  reports  provided  to  senior  management  and  the  board  contain  the   appropriate level of detail and are accurate and timely. The party responsible for this   reporting should assess and report whether the BHC is in compliance with its internal   capital goals and targets and ensure the rationale for any deviations from stated capital   objectives is clearly documented and obtain any necessary approval.

• Some BHCs get an internal audit group to review the data for accuracy and ensure that any   data reported to the board and senior management are given extra scrutiny and crosschecking.   In addition, BHCs with strong practices generally have a strong MIS in place that enables them   to collect, synthesize, analyze, and deliver information quickly and efficiently. These systems   also usually have the ability to run ad hoc analysis to support capital planning as needed   without employing substantial resources. Other BHCs, however, continue to face challenges   with MIS. Many BHCs have systems that are antiquated and/or siloed and not fully compatible, requiring substantial human intervention to reconcile across systems.

• BHCs should have clear and comprehensive documentation for all aspects of  their capital planning processes, including their –

i. Risk-measurement and risk-management infrastructure

ii. Loss and resource estimation methodologies

iii. Process for making capital decision

iv. Efficacy of control and governance functions.

• Documentation should contain sufficient detail, accurately describe BHCs’ practices, allow for review and challenge, and provide relevant information to decision-makers.

Best Practices Regarding Corporate Governance

• BHCs should have strong board and senior management oversight of  their capital planning processes. This includes –

i. Ensuring  periodic  review  of  the  BHC’s  risk  infrastructure  and  loss  and  resource

estimation methodologies

ii. Evaluation of capital goals and targets

iii. Assessment of the appropriateness of stress scenarios considered

iv. Regular review of  any limitations in key processes supporting internal capital planning, such as uncertainty around estimates

v. Approval of capital decisions.

• Together, a BHC’s board and senior management should establish a comprehensive capital   planning process that fits into broader risk-management processes and that is consistent with   the risk-appetite framework and the strategic direction of the BHC.

• A BHC’s board of directors has ultimate oversight responsibility and accountability for capital   planning and should be in a position to make informed decisions on capital adequacy and   capital actions, including capital distributions.

• The board of directors should receive sufficient information to understand the BHC’s material   risks and exposures and to inform and support its decisions on capital adequacy and planning.   The board should receive this information at least quarterly, or when there are material   developments that affect capital adequacy or the manner in which it is assessed.

• Capital adequacy information provided to the board should include capital measures under   current conditions as well as on a post-stress, pro forma basis and should be framed against the   capital goals and targets established by the BHC.

• The information provided to the board should include sufficient details on scenarios used for   the BHC’s internal capital planning so that the board can evaluate the appropriateness of the   scenarios, given the current economic outlook and the BHC’s current risk profile, business   activities, and strategic direction.

• The  information  should  also  include  a  discussion  of  key  limitations,  assumptions,  and   uncertainties within the capital planning process, so that the board is fully informed of any   weaknesses in the process and can effectively challenge reported results before making capital decisions.

• The board should also receive summary information about mitigation strategies to address key   limitations and take action when weaknesses in internal capital planning are identified, applying   additional caution and conservatism as needed.

• BHCs with stronger practices generally have boards that were informed of the risks, exposures,   activities, and vulnerabilities that affected the BHC’s capital adequacy. The board in such a   situation –

i. Understands the major drivers of loss and revenue changes under the scenarios used

ii. Has sufficient expertise and level of engagement to understand and critically evaluate   information provided by senior management.

iii. Recognizes that internal capital planning results are estimates and should be viewed as part of a range of possible results.

iv. Discusses weaknesses identified in the capital planning process, whether they needed to take immediate action to address those weaknesses, and whether the weaknesses are material enough to alter their view of current capital planning results. They also discuss whether a sufficient range of potential stress events and conditions have been considered in assessing capital adequacy.

Best Practices Regarding Corporate Governance-Senior Management

• Senior management is responsible for ensuring that capital planning activities authorized by the   board are implemented in a satisfactory manner and is accountable to the board for the   effectiveness of those activities.

• Senior management should ensure that effective controls are in place around the capital   planning process – including ensuring that the BHC’s stress scenarios are sufficiently severe and   cover the material risks and vulnerabilities facing the BHC. Senior management should make   informed recommendations to the board of  directors about the BHC’s capital, including   capital goals and distribution decisions.

• Senior management should also ensure that proposed capital goals have sufficient analytical   support and fully reflect the expectations of  important stakeholders, including creditors,   counterparties, investors, and supervisors.

• Senior management should identify weaknesses and potential limitations in the capital planning   process and evaluate them for materiality. In addition, it should develop remediation plans for   any weaknesses affecting the reliability of internal capital planning results. Both the specific   identified limitations and the remediation plans should be reported to the board.

• Senior management with stronger practices recognize the imprecision and prevalence of   uncertainty in predicting future outcomes when reviewing information and results from   enterprise-wide scenario analysis.

• At BHCs with stronger practices, senior management maintain an ongoing assessment of all   capital planning areas, identifying and clearly documenting any weaknesses, assumptions,   limitations, and uncertainties, and generally do not consider a one-time assessment of the   capital planning process to be sufficient.

• Furthermore, management should develop clear remediation plans with specific timelines for   resolving identified weaknesses. In some cases, based on its review of the full capital planning   process, senior management make more cautious or conservative adjustments to the capital   plan, such as recommending less aggressive capital actions.

• Management also includes key assumptions and process weaknesses in reports and specifically   points them out to the board, in some cases providing analysis showing the sensitivity of   capital to alternative outcomes.

Best Practices Regarding Corporate Governance-Board Reporting

• The board of directors is required to approve a BHC’s capital plan under the Capital Plan Rule.   In order for boards to carry out this requirement, management should provide adequate   reporting on key areas of the analysis supporting capital plans.

• BHCs with stronger practices generally include information about the independent review and   validation of  models, information on issues identified by internal audit, as well as key   assumptions underpinning stress test results and a discussion of the sensitivity of capital levels   to those assumptions.

• BHCs with stronger practices also supply their boards with information about past capital   planning performance to provide a perspective on how the capital planning process has   functioned over time.

• BHCs with weaker practices generally provide insufficient information to the board of   directors. For example, at some BHCs, capital distribution recommendations do not include all   relevant supporting information and appear to be based on optimistic expectations about how   a given scenario may affect the BHC.

• In addition, the information usually does not specifically identify and address key assumptions   that support the capital planning process. In other cases, the board of directors do not receive   information about governance and controls over internal capital planning, making it difficult to   assess the strength of its capital planning processes and whether results were reliable and   credible.

Best Practices Regarding Corporate Governance-Documenting Decisions

• BHCs should document decisions about capital adequacy and capital actions taken by the   board of directors and senior management, and describe the information used to reach those   decisions.

• Final decisions regarding capital planning of the board or of a designated committee thereof   should be recorded and retained in accordance with the company’s policies and procedures.   BHCs with stronger documentation practices generally have board minutes that describe how   decisions were made and what information was used.

• Documentation can provide evidence that the board challenged results and recommendations,   including reviewing and assessing how senior management challenged the same information.   BHCs with weaker documentation practices generally have board minutes that were very brief   and opaque, with little reference to information used by the board to make its decisions. Some   BHCs do not formally document key decisions.

Best Practices Regarding Capital policy

• A capital policy consists of the principles and guidelines used by a BHC for capital planning, capital issuance, and usage and distributions. A capital policy should include-

i. Internal capital goals

ii. Quantitative or qualitative guidelines for dividends and stock repurchases

iii. Strategies for addressing potential capital shortfalls

iv. Internal governance procedures around capital policy principles and guidelines.

• The capital policy, as a component of a capital plan, must be approved by the BHC’s board of   directors or a designated committee of  the board. It should be a distinct, comprehensive   written  document  that  addresses  the  major  components  of  the  BHC’s  capital  planning   processes and links to and is supported by other policies (risk management, stress testing,   model governance, audit, and others).

• A capital policy should provide details on how a BHC manages, monitors, and makes decisions   regarding  all  aspects  of  capital  planning.  The  policy  should  also  address  –  roles  and   responsibilities of decision-makers, process and data controls, and validation standards.

• Finally, the capital policy should explicitly lay out expectations for the information included in the BHC’s capital plan. A capital policy should describe targets for the level and composition of capital and provide clarity about the BHC’s objectives in managing its capital position.
• The policy should explain how the BHC’s capital planning practices align with the imperative   of maintaining a strong capital position and being able to continue to operate through periods   of severe stress. It should include quantitative metrics such as common stock dividend (and   other) payout ratios as maximums or targets for capital distributions.
• The policy should include an explanation of how management concluded that these ratios are   appropriate, sustainable, and consistent with its capital objectives, business model, and capital   plan. It should also specify the capital metrics that senior management and the board use to   make capital decisions.
• In addition, a capital policy should include governance and escalation protocols that are clear,   credible, and actionable in the event an actual or projected capital ratio target is breached. The   policy should describe processes surrounding how common stock dividend and repurchase   decisions are made and how the BHC arrives at its planned capital distribution amounts.   Specifically, the policy should discuss the following –

i. The main factors and key metrics that influence the size, timing, and form of capital   distributions

ii. The analytical materials used in making capital distribution decisions (e.g., reports, earnings, stress test results, and others)

iii. The specific circumstances that would cause the BHC to reduce or suspend a dividend or stock repurchase program

iv. The factors the BHC would consider if  contemplating the replacement of  common equity with other forms of capital

v. The key roles and responsibilities, including the individuals or groups responsible for   producing the analytical material referenced above, reviewing the analysis, making capital   distribution recommendations, and making the ultimate decisions BHCs should establish   a minimum frequency (at least annually) and other triggers for when its capital policy is   re-evaluated and ensure that these triggers remain relevant and current.

• The capital policy should be re-evaluated and revised as necessary to address changes to   organizational structure, governance structure, business strategy, capital goals, regulatory   environment, risk appetite, and other factors potentially affecting a BHC’s capital adequacy.

• BHCs should develop a formal process for approvals, change management, and documentation   retention relating to their capital policies. Weak capital policies were typically characterized by a   limited scope. They only addressed parts of  the capital planning process, did not provide   sufficient details to convey clearly how capital action decisions will be made, were not well   integrated with or supported by other risk and finance policies, and/or did not contain all of   the  elements  described  above  (e.g.,  clearly  defined  capital  goals,  guidelines  for  capital   distributions and capital composition, etc.).

• In some cases, the capital policy is overly generic and not tailored to the BHC’s unique   circumstances. For example, a policy appearing to be restating supervisory expectations   without concrete examples or BHC-specific considerations. In other cases, the more detailed   procedures are not presented to the board, thus limiting the board’s ability to understand the   analysis underlying its capital planning decisions.

Best Practices Regarding Capital policy-Capital Goals and Targets

• BHCs should establish capital goals aligned with their risk appetites and risk profiles as well as   expectations of internal and external stakeholders, providing specific goals for the level and   composition of capital, both current and under stressed conditions.

• Internal capital goals should be sufficient to allow a BHC to continue its operations during and   after the impact of stressful conditions. As such, capital goals should reflect current and future   regulatory capital requirements, as well as the expectations of shareholders, rating agencies,   counterparties, creditors, supervisors, and other stakeholders.

• BHCs should also establish capital targets above their capital goals to ensure that capital levels   will not fall below the goals during periods of  stress. Capital targets should take into   consideration forward-looking elements related to the economic outlook, the BHC’s financial   condition, the potential impact of stress events, and the uncertainty inherent in the capital   planning process.

• The goals and targets should be specified in the capital policy and reviewed and approved by   the board. In developing their capital goals and targets, particularly with regard to setting the   levels of capital distributions, BHCs should explicitly take into account general economic

• conditions and their plans to grow their on-balance sheet and off-balance sheet size and risks   organically or through acquisitions.

• BHCs should consider the impact of  external conditions during both normal and stressed   economic and market environments and other factors on their overall capital adequacy and   ability to raise additional capital, including the potential impact of contingent exposures and   broader market or systemic events, which could cause risk to increase beyond the BHC’s   chosen risk-tolerance level.

• BHCs should have contingency plans for such outcomes. Additionally, BHCs should calculate   and use several capital measures that represent both leverage and risk, including quarterly   estimates of regulatory capital ratios (including tier 1 common ratio) under both baseline and   stress conditions.

• BHCs with weaker practices in this area generally do not clearly link decisions regarding capital   distributions to capital adequacy metrics or internal capital goals. Weak practices observed in   this area include establishing capital goals based solely on regulatory minimums and the ratios required to be considered well-capitalized without consideration of a BHC’s specific capital needs  given  its  risk  profile,  financial  condition,  business  model  and  strategies,  overall complexity, and sensitivity to changing conditions.
• Some BHCs do not recognize uncertainties and limitations in capturing all potential sources of   losses and in projecting loss and revenue estimates, which reduce the BHCs’ ability to establish   effective capital goals and targets. Other BHCs are not transparent about how they determined   the capital goals and targets in their capital policies.

Best Practices Regarding Capital policy-Capital Contingency Plan

• BHCs should outline in their capital policies, specific capital contingency actions they would   consider to remedy any current or prospective deficiencies in their capital position.

• In particular, a BHC’s policy should include a detailed explanation of the circumstances –   including deterioration in the economic environment, market conditions, or the financial   condition of the BHC – in which it will reduce or suspend a dividend or repurchase program or   not execute a previously planned capital action.

• The policy also should define a set of capital triggers and events that would correspond with   these circumstances. These triggers should be established for both baseline and stress scenarios   and measured against the BHC’s capital targets in those scenarios. These triggers and events   should be used to guide the frequency with which board and senior management will revisit   planned capital actions as well as review and act on contingency capital plans.

• The capital contingency plan should be reviewed and updated as conditions warrant, such as   where there are material changes to the BHC’s organizational structure or strategic direction or   to capital structure, credit quality, and/or market access. Capital triggers should provide an   “early warning” of capital deterioration and should be part of a management decision making framework, which should include target ranges for a normal operating environment and threshold levels that trigger management action.
• Such action should include escalation to the board, potential suspension of capital actions,   and/or activation of a capital contingency plan. Triggers should also be established for other   metrics and events that measure or affect the financial condition or perceived financial   condition of the firm like

i. Liquidity

ii. Earnings

iii. Debt and credit default swap spreads

iv. Ratings downgrades

v. Stock performance

vi. Supervisory actions

vii. General market stress.

• Contingency actions should be flexible enough to work in a variety of  situations and be   realistic for what is achievable during periods of stress. The capital plan should be prepared   recognizing that certain capital-raising and capital-preserving activities may not be feasible or   effective during periods of stress.

• BHCs should have an understanding of market capacity constraints when evaluating potential   capital actions that require accessing capital markets, including debt or equity issuance and also   contemplated asset sales. Contingency actions should be ranked according to ease of execution   and their impact and should incorporate the assessment of stakeholder reactions (e.g., impacts   on future capital-raising activities).

• Weak capital contingency plans provide few options to address contingency situations and/or   do  not  consider  the  feasibility  of  options  under  stressful  conditions.  Plans  with  overly   optimistic assumptions or excessive reliance on past history (in terms of  both possible   contingency situations and options to address those situations) are also considered weak, as are   plans that lack support for the feasibility and availability of possible contingency actions.

• Other weak practices include-

i. Establishing triggers based on actual results but not on projected results.

ii. Establishing  triggers  based  on  minimum  regulatory  capital  ratios  only  with  no   consideration  of  the  expectations  of  other  stakeholders  including  counterparties,   creditors and investors, or of other metrics or market indicators.

Best Practices Stress Testing And Scenario Design

• Under the Capital Plan Rule, a BHC is required to use a BHC developed stressed scenario that   is appropriate for its business model and portfolios. Accordingly, BHCs should have a process   for designing scenarios for enterprise-wide scenario analysis that reflects the BHC’s unique   business activities and and associated vulnerabilities.

• The range of observed practice for developing BHC stress scenarios is broad. Some BHCs   design stress scenarios using internal models and expertise. Other BHCs use vendor-defined   macroeconomic  scenarios  or  use  vendor  models  to  define  customized  macroeconomic   scenarios.

• For BHCs with internally developed scenarios, those with stronger scenario-design practices   use internal models in combination with expert judgment rather than relying solely on either   models or expert judgment to define scenario conditions and variables.

• Among BHCs that use third-party scenarios, those with stronger practices tailored third-party-defined scenarios to their own risk profiles and unique vulnerabilities.

• Regardless of the method used to develop the scenario, BHCs should have a scenario-selection process that engages a broad range of  internal stakeholders such as risk experts, business managers, and senior management.
• Although they are required to submit only one BHC stress scenario for 𝐶𝐶𝐴𝑅, BHCs should   develop a suite of scenarios that collectively capture their material risks and vulnerabilities   under a variety of stressful circumstances and should incorporate them into their overall capital   planning processes.
• As indicated in the preamble to the Capital Plan Rule, “the bank holding company-designed stress   scenario should reflect an individual company’s unique vulnerabilities to factors that affect its firm-wide activities   and risk exposures, including macroeconomic, market-wide, and firm-specific events”. Thus, BHC stress   scenarios should reflect macroeconomic and financial conditions that are tailored specifically to   stress a BHC’s key vulnerabilities and idiosyncratic risks, based on factors such as its particular business model, mix of assets and liabilities, geographic footprint, portfolio characteristics, and   revenue drivers.
• Thus, BHCs with stronger scenario-design practices clearly and creatively tailor their BHC   stress scenarios to their unique business-model features, emphasizing important sources of risk not captured in the supervisory severely adverse scenario. Examples of such risks observed in practice include a significant counterparty default; a natural disaster or other operational-risk event; and a more acute stress on a particular region, industry, and/or asset class as compared to the stress applied to general macroeconomic conditions in the supervisory adverse and severely adverse scenarios.
• The set of variables that a BHC includes in its stress scenario should be sufficient to address all   material risks arising from its exposures and business activities. A business line can face   significant stress from multiple sources, requiring more than one risk factor or macroeconomic   variable.
• The scenario should generally contain the relevant variables to facilitate pro forma financial   projections that capture the impact of changing conditions and environments. BHCs should   have a consistent process for determining the final set of variables and provide this rationale as   part of the scenario narrative.
• Overall, BHCs with stronger scenario-design practices generate scenarios in which the link   between the variables included in the scenario and sources of risk to the BHC’s financial outlook are transparent and straightforward. Clear narratives help make these links more transparent. BHCs with weaker scenario design practices develop stress scenarios that exclude some variables relevant to the BHC’s risk profile and idiosyncratic vulnerabilities. For example, some BHCs with significant trading activities and revenues include a limited set of relevant financial variables. Other BHCs with significant regional and/or industry concentrations do not include relevant geographic or industry variables.
• Scenarios should be supported by a clear narrative describing how the scenario addresses the   particular vulnerabilities and material risks facing the BHC. BHCs with stronger scenario   design practices provide narratives describing how the scenario variables related to the risks faced by a BHC’s significant business lines and, in some cases, how the scenario variables   correspond to variables in the BHC’s internal risk-management models.
• The  narratives  also  provide  explanations  of  how  a  scenario  stressed  a  BHC’s  unique   vulnerabilities specific to its business model and how the paths of the scenario variables related   to each other in an economically intuitive way.
• Weaker practices include scenario narratives that do not provide any context for the variable paths as well as scenario narratives that describe features that are not reflected in any variables considered in a BHC’s internal capital planning.

Estimation Methodologies For Losses, Revenues & Expenses

• A BHC’s capital plan must include estimates of projected revenues, expenses, losses, reserves,   and pro forma capital levels, including any minimum regulatory capital ratios, the tier 1 common ratio and any additional capital measures deemed relevant by the BHC, over the   planning horizon under expected conditions and under a range of stressed scenarios.

• Projections of losses, revenues, and expenses under hypothetical stressed conditions serve as   the fundamental building blocks of the pro forma financial analysis supporting enterprise-wide   scenario analysis. BHCs should have stress testing methodologies that generate credible   estimates that are consistent with assumed scenario conditions.

• It is important for BHCs to understand the uncertainties around their estimates, including the   sensitivity of the estimates to changes in inputs and key assumptions. Overall, BHCs’ estimates of losses, revenues, and expenses under each of  the scenarios should be supported by   empirical evidence, and the entire estimation process should be transparent and repeatable. The   Federal Reserve generally expects BHCs to use models or other quantitative methods as the basis for their estimates; however, there may be instances where a management overlay or other   qualitative approaches may be appropriate due to data limitations, new products or businesses, or other factors.

Quantitative Basis For Enterprise-Wide Scenario Analysis

• Generally, BHCs should develop and use internal data to estimate losses, revenues, and   expenses as part of enterprise-wide scenario analysis. However, in certain instances, it may be   more appropriate for BHCs to use external data to make their models more robust. For   example, BHCs may lack sufficient, relevant historical data due to factors such as systems   limitations, acquisitions, or new products.

• When using external data, BHCs should take care to ensure that the external data reasonably   approximates underlying risk characteristics of  their portfolios and make adjustments to   modeled outputs to account for identified differences in risk characteristics and performance   reflected in internal and external data.

• BHCs can use a range of quantitative approaches to estimate losses, revenues, and expenses,   depending on the type of portfolio or activity for which the approach is used, the granularity   and length of available time series of data, and the materiality of a given portfolio or activity.

• While the Federal Reserve does not require BHCs to use a specific estimation method, each   BHC should estimate its losses, revenues, and expenses at sufficient granularity so that it can   identify common, key risk drivers and capture the effect of changing conditions and environments.

• While BHCs often segment their portfolios and activities along functional areas, such as by line   of business or product type, the leading practice is to determine segments based on common   risk  characteristics  (e.g.,  credit  score  ranges  or  loan-to-value  ratio  ranges)  that  exhibit   meaningful differences in historical performance.

• The granularity of segments typically depends on the type, size, and composition of the BHC’s   portfolio.  For  example,  a  more  diverse  portfolio  –  both  in  terms  of  borrower  risk   characteristics and performance – would generally require a greater number of segments to   account for the heterogeneity of the portfolio.

• As a general practice, BHCs should separately estimate losses, revenues, or expenses for   portfolios or business lines that are sensitive to different risk drivers or sensitive to risk drivers   in a markedly different way.

• BHCs with leading practices generally demonstrate clearly the rationale for selecting certain risk   drivers over others. BHCs with lagging practices generally use risk drivers that do not have a clear link to results, either statistically or conceptually.

• Many models used for stress testing require a significant number of assumptions to implement.   Further, the relationship between macroeconomic variables and losses, revenues, or expenses   could differ considerably in the hypothetical stress scenario from what is observed historically.   As a result, while traditional tools for evaluating model performance (such as comparing   projections to historical out-of-sample outcomes) are still useful, the Federal Reserve expects   BHCs to supplement them with other types of analysis.

• Sensitivity analysis is one tool that some BHCs have used to test the robustness of models and   to help model developers, BHC management, the board of directors, and supervisors identify   the assumptions and parameters that materially affect outcomes. Sensitivity analysis can also   help ensure that core assumptions are clearly linked to outcomes.

• Using results from different estimation approaches (challenger models) as a benchmark is   another way BHCs can gain greater comfort around their primary model estimates, as the   strengths of one approach could potentially compensate for the weaknesses of another. When   using multiple approaches, however, it is important that BHCs have a consistent framework for evaluating the results of different approaches and supporting rationale for why they chose the methods and estimates they ultimately used.
• In certain instances, BHCs may need to rely on third-party models for reasons like limitations   in internal modeling capacity. In using these third-party models (vendor models or consultant-   developed models), BHCs should ensure that their internal staff have working knowledge and   a good conceptual understanding of the design and functioning of the models and potential   model limitations so that management can clearly communicate them to those governing the   process.
• An off  the shelf  vendor model often requires some level of  firm-specific analysis and   customization  to  demonstrate  that  it  produces  estimates  appropriate  for  the  BHC  and   consistent with scenario conditions.
• Sensitivity analysis can be particularly helpful in understanding the range of possible results of   vendor models with less transparent or proprietary elements. Importantly, all vendor and   consultant-developed models should be validated in accordance with SR 11-7 guidelines.

• Some BHCs generate annual projections for certain loss, revenue, or expense items and then   evenly distribute them over the four quarters of each year. This practice does not reflect a   careful estimate of the expected quarterly path of losses, net revenue, and capital, and thus is   only acceptable when a BHC can clearly demonstrate that the projected item is highly   uncertain, and the practice likely results in a conservative estimate.

Qualitative Projections, Expert Judgements And Adjustments

• While quantitative approaches are important elements of enterprise-wide scenario analysis,   BHCs should not rely on weak or poorly specified models simply to have a modeled approach.   In fact, most BHCs use some forms of expert judgment for many purposes- generally as a   management adjustment overlay to modeled outputs.

• BHCs can, in limited cases, also use expert judgment as the primary method to produce an   estimate of losses, revenue, or expenses. BHCs may use a management overlay to account for   the unique risks of certain portfolios that are not well captured in their models, or otherwise to   compensate for specific model and data limitations.
• In using expert judgment, BHCs should ensure that –

i.They have a transparent and repeatable process

ii.Management judgments are well supported

iii.Key assumptions are consistent with assumed scenario conditions.

• As  with  quantitative  methods,  the  assumptions  and  processes  that  support  qualitative   approaches should be clearly documented so that an external reviewer can follow the logic and   evaluate the reasonableness of the outcomes.

• BHCs should evaluate a range of potential estimates and conduct sensitivity analysis for key   assumptions used in the estimation process.

• Extensive use of management judgment to adjust modeled estimates should trigger review and   discussion as to whether new or improved modeling approaches are needed. In reporting to   the board of directors, management should always provide both the initial results and the   results after any judgmental adjustments.

Loss Estimation Methodologies-Retail & Wholesale Credit Risks

• Sources of data used for loss estimation have often differed between retail and wholesale   portfolios. Due to availability of a richer set of retail loss data, particularly from the most   recent downturn, BHCs generally use internal data to estimate defaults or losses on retail   portfolios and only infrequently use external data with longer history to benchmark estimated   losses on portfolios that had more limited loss experience in the recent downturn.

• For wholesale portfolios, some BHCs supplemented internal data with external data or used   external data to calibrate their models due to a short time series (5-10 years) that included only   a single downturn cycle.

• BHCs with stronger practices account for dynamic changes in their portfolios, such as loan   modifications or changes in portfolio risk characteristics and make appropriate adjustments to   data or estimates to compensate for known data limitations (including lack of historical periods   of stress). BHCs with weaker practices generally fail to compensate for data limitations or   adequately demonstrate that external data reasonably reflect the BHC’s actual exposures, often   failing to capture geographic, industry, or lending-type concentrations.

• The level of segmentation used for modeling varies, depending on the type and size of the portfolio and estimation methods used. For example, BHCs often segment the retail portfolio based on – some combinations of product, the lien position, the risk characteristics (such as credit score, loan-to-value ratio, and collateral), and underlying collateral information (e.g., single-family home versus condominium) with some models being estimated at the loan-level and others at the portfolio level.
• BHCs with stronger practices have segmentation schemes that are well supported by the BHC’s   data and analysis, with sufficient granularity to capture exposures that react differently to risk   drivers under stressed conditions.
• BHCs with weaker practices generally use a single model for multiple portfolios, without   sufficiently  adjusting  modeling  assumptions  to  capture  the  unique  risk  drivers  of  each   portfolio. For example, in estimating losses on wholesale portfolios, these BHCs do not   adequately allow for variation in loss rates commonly attributed to industry, obligor type,   collateral, lien position, or other relevant information.

Loss Estimation Methodologies-Common Credit Loan Loss

• BHCs have used a wide range of methods to estimate credit losses, depending on the type and   size of portfolios and data availability.

• These methods can be based on either an accounting-based loss approach (charge-off  and   recovery) or an economic loss approach (expected losses). BHCs have flexibility in selecting a   specific  loss  or  estimation  approach,  however,  it  is  important  for  BHCs  to  understand   differences between the two loss approaches, particularly in terms of  the timing of  loss   recognition, and to account for the differences in setting the appropriate level of reserves at the end of each quarter

Common Credit Loan Loss-Expected Loss Approaches

• Under the expected loss approach, losses are estimated as a function of three components –   probability of default (𝑃𝐷), loss given default (𝐿𝐺𝐷), and exposure at default (𝐸𝐴𝐷). 𝑃𝐷, 𝐿𝐺𝐷, and 𝐸𝐴𝐷 can be estimated at a segment level or at an individual loan level, and using different models or assumptions.

• In general, BHCs use econometric models to estimate losses under a given scenario, where the   estimated 𝑃𝐷𝑠 are conditioned on the macroeconomic environment and portfolio or loan   characteristics.

• BHCs with leading practices are generally able to break down losses into 𝑃𝐷, 𝐿𝐺𝐷, and 𝐸𝐴𝐷   components, separately identifying key risk drivers for each of those components, though they   typically do not demonstrate this level of granularity consistently across all portfolios.

• By design, estimates based on long-run average behavior over a mix of conditions (including   periods of economic expansion and downturn) are not appropriate for projecting losses under   stress and should not be used for these purposes.

• BHCs with leading practices usually tie 𝐿𝐺𝐷 to underlying risk drivers, accounting for collateral and guarantees, and also incorporating the likelihood of a decline in collateral values under stress.
• Most BHCs have more limited data on 𝐿𝐺𝐷 and, as a result, BHCs often applied a simple,   conservative assumption based stressed 𝐿𝐺𝐷 on their experience during the crisis or scale up   the historical average 𝐿𝐺𝐷 using expert judgment.
• In using such methods, it is important for BHCs to ensure that the process is well supported,   transparent and in line with the Federal Reserve’s general expectation for expert judgment-   based estimates.
• Wherever possible, BHCs should benchmark their estimates with external data or research and   analysis. BHCs with lagging practices generally model 𝐿𝐺𝐷 using a weighted average approach   at an aggregate portfolio level, without some level of segmentation (e.g., by lending product,   priority of claim, collateral type, geography, vintage).
• BHCs may fail to demonstrate that 𝐿𝐺𝐷 estimates are consistent with the severity of  the   scenario. Although some BHCs find a relationship between 𝐸𝐴𝐷 and credit quality, most BHCs do not model 𝐸𝐴𝐷𝑠 to vary according to the macroeconomic environment, in large part due to data limitations. Rather, many BHCs apply a static assumption to estimate stressed 𝐸𝐴𝐷.
• BHCs with stronger practices generally include the use of loan equivalent calculations (i.e.,   estimated additional draw-downs as a percentage of unused commitments, which are added to   the outstanding or drawn balance) and credit-conversion factors (i.e., additional drawdowns   during the period leading up to default – usually one year prior – as a percentage of both drawn   and undrawn commitments) to capture losses associated with undrawn commitments.
• BHCs with weaker practices usually do not project stressed exposures associated with undrawn   commitments or rely on the assumption that they can actively manage down committed lines   during stress scenarios.

• Sound rating transition models require two fundamental building blocks –

a)A robust time series of data

b)A well-calibrated, granular-risk rating systems

• The Federal Reserve expects BHCs, that use rating transition models, to have robust time series   of data that include a sufficient number of transitions, which allows BHCs to establish a   statistically  significant  relationship  between  the  transition  behavior  and  macroeconomic   variables.

• Data availability has been a widespread constraint inhibiting the development of  granular   transition models because a sufficient number of upgrades and downgrades are necessary to   preclude sparse matrices. In order to overcome these data limitations, BHCs have often relied   on third-party data to develop rating transition models. Consistent with the Federal Reserve’s   general expectations, when using third-party data, BHCs should be able to demonstrate that   the transition matrices estimated with external data are a reasonable proxy for the migration   behavior of their portfolios.

• Rating transition models also require granular ratings systems that capture differences in the potential  for  defaults  and  losses  for  a  given  set  of  exposures  in  various  economic environments. BHCs with stronger practices typically have more granular ratings system and account for limitations in their data and credit rating systems by making adjustments to model assumptions or estimates, or by supplementing internal data with external data. BHCs with weaker practices often fail to demonstrate that supplemented external data adequately reflects the ratings performance of the BHC’s portfolio.
• BHCs with weaker practices also sometimes rely on a risk rating process that historically   resulted in lumpiness in rating upgrades and downgrades or material concentrations in one or two rating categories. As a result, these BHCs often produced transition matrices with limited   sensitivity to scenario variables, and resulting estimates were more consistent with long-term   average default rates than with default rates that would be experienced under severe economic stress.

Common Credit Loan Loss-Roll Rate Models

• Many BHCs use roll-rate models to estimate losses for various retail portfolios. Roll-rate   models generally estimate the rate at which loans that are current or delinquent in a given   quarter roll into delinquent or default status in the next period.

• As a result, these models are conceptually similar to rating transition models. The Federal   Reserve expects BHCs that use roll-rate models to have a robust time series of data with   sufficient granularity. The robust time series data allow the BHC to establish a strong   relationship between roll rates and scenario variables, while the availability of granular data   enables BHCs to model all relevant loan transitions and to segment the portfolio into sub   portfolios that exhibit meaningful variations in performance, particularly during the period of   stress.

• In general, BHCs should estimate roll rates using models that are conditioned on scenario   variables. For certain transition states where statistical relationships between roll rates and   scenarios are weak (such as late-stage loan delinquency), BHCs should incorporate conservative   assumptions rather than relying solely on statistical relationships.

• While roll-rate models have some advantages, including transparency and ease of use, they often have a weak predictive power outside the near future, particularly if  they are not properly conditioned on scenario variables.
• As a result, some roll-rate models have limited usefulness for stress testing over a longer   horizon, such as the nine-quarter planning horizon required in 𝐶𝐶𝐴𝑅. Some BHCs have used   roll-rate models in conjunction with other estimation approaches (such as a vintage model) that   project losses for later periods.
• In general, it is a weaker practice to combine two different models, as it can introduce   unexpected jumps in estimated losses over the planning horizon, though some BHCs have   judgmentally weighed two different estimation methods to smooth projected losses.
• If BHCs combine two models, they should be able to demonstrate that such an approach is   empirically warranted based on output analysis, including sensitivity analysis, and that the   process of transitioning from one set of results to the other is consistent, well supported, and   transparent.

Common Credit Loan Loss-Vintage Loss Models

• Some BHCs use vintage loss models, also known as age-cohort time models, to estimate losses   for certain retail portfolios. BHCs that use vintage loss models generally segment their retail   portfolios by vintage and collateral – or credit-quality-based segments.

• Losses are estimated using a multistep process – developing a baseline seasoning curve for each   segment and using a regression model to estimate sensitivity of  losses to macroeconomic   variables at each seasoning level (e.g., four quarters after origination).

• This technique is commonly used in several vendor models, but BHCs also have developed and   used proprietary models using this technique. These models have several advantages (such as   natural segmentation of portfolio by cohort and maturity) and ease of application to credit   products (such as auto loans) that exhibit lifecycle effects.

• However, vintage models can be very challenging to construct, calibrate, and validate. In   particular, it may be difficult to separately identify vintage effects from the effects of   macroeconomic variables, which can result in poorly specified models.

• These models also assume that different cohorts will experience similar losses over time, generating results that are representative of average years, rather than during the period of stress. In using vintage models, it is important for a BHC to be able to demonstrate that the approach appropriately reflects its portfolio composition and history, and that modeled outputs are consistent with stressed conditions

Common Credit Loss-Charge Off Models

• A minority of  BHCs have used net charge-off  (𝑁𝐶𝑂) models as either a primary loss-   estimation model or a benchmark model. Typically, the 𝑁𝐶𝑂 models BHCs use, estimate a   statistical relationship between charge-off rates and macroeconomic variables at a portfolio   level, and often includes autoregressive terms (lagged 𝑁𝐶𝑂 rates).

• While some BHCs also incorporate variables that describe the underlying risk characteristics of   the portfolio, 𝑁𝐶𝑂  models that BHCs use for capital planning generally do not capture   variation in sensitivities to risk drivers across important portfolio segments nor account for   changes in portfolio risk characteristics over time.

• As a matter of general practice, BHCs should not use models that do not capture changes in   portfolio risk characteristics over time and in scenarios used for stress testing as part of their   internal capital planning.

• 𝑁𝐶𝑂  models  often  exhibit  lower  explanatory  power  than  models  that  consider  distinct   portfolio risk drivers. In addition, 𝑁𝐶𝑂 models implicitly assume that historical charge-off   performance  is  a  good  predictor  of  future  performance,  when  in  reality,  the  historical   relationship between charge-offs and macro variables may not be realized under very stressful scenarios that fall outside the portfolio’s actual historical experience.
• Accordingly, a 𝑁𝐶𝑂 model that is estimated without using sufficient segmentation or does not   account for current or changing portfolio composition is unlikely to produce robust loss   estimates. Thus, BHCs should avoid using such a 𝑁𝐶𝑂 model as the primary loss-estimation   approach for a material portfolio.

Common Credit Loan Loss-Scalar Adjustments

• Some BHCs have used simple scalars to adjust portfolio loss estimate, under a baseline scenario, upward for stress scenarios.

• Scalars have been calibrated based on some combination of historical performance, the ratio   of  modeled stressed losses to baseline losses estimated for other portfolios, and expert   judgment.

• Scalar adjustments are easy to develop, implement, and communicate. However, the approach has significant shortcomings including lack of transparency and lack of sensitivity to changes   in portfolio composition and scenario variables.

• Consequently, the use of these types of approaches should be, at most, limited to immaterial portfolios.

Operational Risk

• Best practices in operational-risk models are still evolving, and the Capital Plan Rule does not   require BHCs to use advanced measurement approach (𝐴𝑀𝐴) models for stressed operational   risk loss estimation.

• However, BHCs that have developed a rich set of data to support the 𝐴𝑀𝐴 should consider   leveraging the same data and risk-management tools to estimate operational losses under a   stress scenario, regardless of a particular methodology they choose to estimate losses.

• Most operational-risk models use historical data on operational risk loss “events“ – incidences   in which a BHC has experienced a loss or been exposed to loss due to inadequate or failed   internal processes, people, or systems or from external events.

• Generally, operational-risk events are grouped into one of several event type categories, such as   internal fraud, external fraud, or damage to physical assets. In general, BHCs should use   internal operational-loss data as a starting point to provide historical perspective, and then   incorporate forward-looking elements, idiosyncratic risks, and tail events to estimate losses.

• Most BHCs have supplemented their internal loss data with external data when modeling operational-risk loss estimates and scaled the losses to make the external loss data more commensurate with their individual risk profiles. The Federal Reserve expects such scaling approaches to be well supported.

• Few BHCs have incorporated business environment and internal control factors such as risk   control self-assessments and other risk indicators into their operational-risk methodology.   While the Federal Reserve does not expect BHCs to use these qualitative tools as direct inputs   in a model, they can help identify areas of potential risk and help BHCs select appropriate   scenarios that stress those risks.

Operational Risk-Internal Data Collection & Data Quality

• The Federal Reserve expects BHCs to have a robust and comprehensive internal data-   collection method that captures key elements, such as critical dates (i.e., occurrence, discovery,   and accounting), event types, and business lines. In general, BHCs should use complete data   sets of internal losses when modeling, and not judgmentally exclude certain loss data.

• Data quality and comprehensiveness have varied considerably across BHCs. BHCs with lagging   practices often exclude certain internal loss data from model input for various reasons.   Examples include-

i. Excluding large items such as legal reserves and tax/compliance penalties

ii. Omitting losses from merged or acquired institutions mergers or acquisitions due to complications in collection and aggregation

iii. Excluding loss data from discontinued business lines, even though the loss events were

reasonably generic and applicable to remaining business lines within the organization.

• Some BHCs have addressed observed outliers by –

i. Omitting them from the data set

ii. Modeling them separately

iii. Applying an addon based on scenario analysis or management input

• If BHCs do not have the data from potential mergers and acquisitions, one way to account for   this limitation is to scale existing internal data using the size of operations and apply an add-on   to applicable business lines or units of measure.

• If a BHC excludes data or uses data-smoothing techniques, especially as they affect large losses,   it should have a well-supported rationale for doing so. This should also be clearly documented.   The Federal Reserve expects BHCs to segment their loss data into units of measure that are   granular enough to capture similar losses while balancing it with the availability of data.

• Most BHCs have segmented datasets by event type, however, some BHCs have segmented the   loss data by consolidated business lines, event types, or some combination of the two.

Operational Risk-CorrelationWith Macroeconomic Factors

• Most BHCs have attempted to identify correlation between macroeconomic factors and   operational-risk losses, but some have struggled to identify a clear relationship for some types   of operational-risk loss events.

• BHCs that did not identify a significant correlation typically developed other methodologies,   such as scenario analysis layered onto modeled results, to project stressed operational-risk   losses. These approaches can be reasonable alternatives if BHCs can demonstrate that their   approach results in sufficiently conservative loss estimates that are consistent with the stress   scenario.

• BHCs  that  identified  correlations  between  macroeconomic  factors  and  operational-risk   elements typically had large data sets and often used external loss data to supplement internal   data. These BHCs often identified correlations between loss frequency and macroeconomic   factors for certain event types and adjusted the frequency distributions for the respective event type accordingly.

Operational Risk-Common Operational Loss Estimates Approach

• Most BHCs have used their annual budgeting or forecasting process to estimate operational   losses in the baseline scenario. The process typically uses a combination of historical loss data   and management input at a business-line level.

• Some BHCs have used historical averages from internal loss data to estimate losses in the   baseline  scenario.  BHCs  with  stronger  practices  use  a  combination  of  approaches  to   incorporate historical loss experience, forward-looking elements, and idiosyncratic risks into   their stressed loss projections.

• Using a combination of approaches can help address model and data limitations. Some BHCs   use separate models for certain events types such as fraud or litigation, and use other   approaches  (e.g.,  using  historical  averages)  for  event  types  where  no  correlation  with   macroeconomic factors was identified.

• A simple approach may be acceptable depending on the size and complexity of the BHC as   well as data and sophistication of models available to them. Very few BHCs have yet developed   benchmarks to either challenge or further support the projections provided by their main   models.

Operational Risk-Regression Models

• Most BHCs have used a regression model, either by itself or with another approach described   below, to estimate operational risk losses for stress scenarios. Some BHCs use regression   models for the baseline scenarios, albeit with different parameters. Operational-risk regression   models are generally used to estimate two variables: loss frequency (i.e., the number of   operational-risk losses) and loss severity (i.e., the loss amount).

• BHCs that are able to identify significant correlation between macroeconomic variables and   operational-risk  losses  can  use  regression  models  to  stress  the  loss  frequency  or  total   operational-risk losses.

• Some macroeconomic variables can be adjusted for the purpose of correlation analysis or to   reflect time-lag assumptions. Most BHCs judgmentally choose time periods for estimation and   model specification rather than justifying them with statistical evidence.

• Most BHCs are not able to find meaningful correlation between macroeconomic variables and   operational-risk loss severity. As a result, BHCs that use a regression model to estimate loss   frequency typically applies the loss-severity assumption (e.g., static or four-quarter moving   average) based on the most recent crisis period to estimate operational losses.

Operational Risk-Modified Loss Distribution Approach

• The 𝐿𝐷𝐴 is an empirical modeling technique commonly used by BHCs subject to the 𝐴𝑀𝐴 to   estimate annual value-at-risk (𝑉𝑎𝑅) measures for operational-risk losses based on loss data and   fitted parametric distributions. The 𝐿𝐷𝐴 involves estimating probability distributions for the   frequency and the severity of operational loss events for each defined unit of measure, whether   it is a business line, an event type, or some combination of the two.

• The estimated frequency and severity distributions are then combined, generally using a Monte   Carlo simulation, to estimate the probability distribution for annual operational-risk losses at   each unit of measure.

• For purposes of 𝐶𝐶𝐴𝑅, 𝐿𝐷𝐴 models have generally been used in one of two ways –

i.By using a lower confidence interval than the 99.9^th percentile used by the 𝐴𝑀𝐴

ii.By adjusting the frequency based on outcomes of correlation analysis.

• BHCs that modified the 𝐿𝐷𝐴 by using a lower confidence interval typically have used either the   mean or median for the baseline estimates and higher confidence intervals – typically ranging from 70^th percentile to 98^th percentile – for the stressed estimates.

• Additionally, some BHCs have used different confidence intervals for different event types.   The Federal Reserve does not require BHCs to use a particular percentile to produce stressed   estimates. However, it expects BHCs to –

1. Implement a credible, transparent process to select a percentile

2. Be able to demonstrate why the percentile is an appropriate choice given the specific scenario under consideration

3. Perform sensitivity analyses around the selection of a percentile to test the impact of this assumption on model outputs.

• Some BHCs modified the 𝐿𝐷𝐴 by adjusting frequency distributions based on the observed   correlation between macroeconomic variables and operational risk losses.

Operational Risk-Scenario Analysis

• Scenario analysis is a systematic process of obtaining opinions from business managers and   risk-management  experts  to  assess  the  likelihood  and  loss  impact  of  plausible  severe   operational loss events.

• Some BHCs have used this process to determine a management overlay that is added to losses   estimated  using  a  model-based  approach.  BHCs  have  used  this  overlay  to  incorporate   idiosyncratic risks (particularly for event types where correlation was not identified) or to   capture potential loss events that the BHC had not previously experienced.

• BHCs should be able to demonstrate the quantitative effect of the management overlay on   final loss estimates. Scenario analysis, if used effectively, can help compensate for data and   model limitations, and allows BHCs to capture a wide range of risks, particularly where limited   data is available.

• The Federal Reserve expects BHCs using scenario analysis to have a clearly defined process   and provide an appropriate rationale for the specific scenarios included in their loss estimate.   The process for choosing scenarios should be credible, transparent, and well supported.

Operational Risk-Historical Averages

• Some BHCs use historical averages of  operational-risk losses, in combination with other   approaches noted before, to estimate operational-risk losses under stress scenarios.

• BHCs  have  used  historical  averages  for  event  types  where  no  correlation  between   macroeconomic factors and operational risk losses was identified but used a regression model   for event types where correlations were identified.

• A small number of  BHCs have used historical averages as the sole approach to develop   stressed loss estimates. When used alone, this approach is backward-looking and excludes   potential risks the BHCs have not experienced.

• When using historical averages, BHCs should support the chosen time periods, thresholds, and   any excluded or adjusted outliers and demonstrate that loss estimates are consistent with what   are expected in the stress scenario.

Market Risk And Counterparty Credit Risk

• BHCs that have sizeable trading operations may incur significant losses from such operations   under a stress scenario due to valuation changes stemming from credit or market risk, which   may arise as a result of moves in risk factors such as interest rates, credit spreads, or equity and   commodities prices, and counterparty credit risk owing to potential deterioration in the credit   quality or outright default of a trading counterparty.

• BHCs use different techniques for estimating such potential losses. These techniques can be broadly grouped into two approaches-

i.Probabilistic  approaches  that  generate  a  distribution  of  potential  portfolio-level

profit/loss (𝑃/𝐿)

ii.Deterministic approaches that generate a point estimate of portfolio-level losses under a specific stress scenario.

• Both  approaches  have  different  strengths  and  weaknesses.  A  probabilistic  approach  can provide useful insight into a range of  scenarios that generate stress losses in ways that a deterministic  stress  testing  approach  may  not  be  able  to  do.  However,  the  probabilistic approach is complex and often lacks transparency, and as a result, it can be difficult t communicate the relevant scenarios to senior managers and the board of directors.

• In addition, the challenges inherent in tying probabilistic loss estimate to specific underlying   scenarios can make it difficult for management and the board of directors to readily discern   what actions could be taken to mitigate portfolio losses in a given scenario.

• Combined, these factors complicate the use of probabilistic approaches as the primary element   in  an  active  capital  planning  process  that  reflects  well-informed  decisions  by  senior   management and the board of directors.

• The Federal Reserve expects BHCs using a probabilistic approach to provide evidence that   such an approach can generate scenarios that are potentially more severe than what was   historically experienced, and also to clearly explain how BHCs use the scenarios associated with   tail losses to identify and address their idiosyncratic risks.

• By comparison, a deterministic approach generally produces scenarios that are easier to   communicate to senior management and the board of  directors. However, a deterministic   approach often uses a limited set of scenarios and may miss certain scenarios that may result in   large losses.

• The Federal Reserve expects BHCs using a deterministic approach to demonstrate that they   have considered a range of scenarios that sufficiently stress their key exposures. For 𝐶𝐶𝐴𝑅,   most BHCs generally relied on a deterministic approach. BHCs using deterministic approaches   often  relied  on  statistical  models,  for  example,  to  inform  the  magnitude  of  risk-factor   movements and covariances between risk factors, and also considered multiple scenarios as   part of the broader internal stress testing supporting their capital planning process.

• BHCs using deterministic approaches used a three-step process to generate 𝑃/𝐿 losses under a stress scenario-

i.Design and selection of stress scenarios

ii.Construction  and  implementation  of  the  scenario  (that  is,  translation  to  risk-factor moves)

iii.Revaluation  (and  aggregation)  of  position  and  portfolio  level  𝑃&𝐿  under  the  stress scenarios

• The Federal Reserve expects BHCs to have robust operational and implementation practices in all areas, including position inclusion, risk-factor representations, and revaluation methods.

Impact of capital Adequacy-Balancesheet & Rwas

• BHCs should have a well-documented process for generating projections of  the size and   composition of on-balance sheet and off-balance sheet positions and Risk Weighted Assets   (𝑅𝑊𝐴) over the scenario horizon. Balance projections are a key input to enterprise-wide   scenario analysis given their direct impact on the estimation of losses, 𝑃𝑃𝑁𝑅, and 𝑅𝑊𝐴.

• Estimating the evolution of balance sheet size and composition under stress integrates many   interrelated features. For example, loan balances and the stock of 𝐴𝐹𝑆 securities at a point in   time will depend upon origination, purchase, and sale activity from period to period, as well as   maturities, prepayments, and defaults.

• Due to complexities related to dynamically projecting and integrating various components (e.g.,   originations, prepayments and defaults), most BHCs made direct projections of balances for   each major segment of the balance sheet (e.g., loans, deposits, trading assets and liabilities, and   other assets) for each quarter of the scenario horizon.

• BHCs often face challenges in integrating the ultimate balance projections with other aspects   (for example, borrower or depositor behavior). BHCs with stronger practices separately   consider the drivers of change to asset and funding balances, such as contractual paydowns, modeled prepayments, nonperformance, and new business activity for assets, rather than simply projecting targeted balances directly.
• At these BHCs, each element is separately assessed for consistency with scenario conditions   and other management assumptions. BHCs with stronger practices also either directly consider   the impact of these various factors in their balance projections or have procedures to evaluate   the reasonableness of any implied behavior by including input from business line leaders in the   process and iterating to reasonable estimates in a well-supported and transparent manner.
• BHCs should clearly establish and incorporate into their scenario analysis the relationships   among and between revenue, expense, and on-balance-sheet and off-balance-sheet items under   stressful conditions.
• Most BHCs use asset-liability management (𝐴𝐿𝑀) software as a part of their enterprise-wide   scenario-analysis toolkit, which helps integrate these items. BHCs that do not use 𝐴𝐿𝑀   software must have a process that integrates balance sheet projections with revenue, loss, and   new business projections.
• BHCs  with  more  tightly  integrated  procedures  are  better  able  to  ensure  appropriate   relationships among the scenario conditions, losses, expenses, revenue, and balances. As noted   above, BHCs should not rely on favorable assumptions that cannot be reasonably assured in   stress scenarios given the high level of uncertainty around market conditions.

• Examples of aggressive or favorable balance sheet assumptions include-

i. Large changes in asset mix that serve to decrease BHCs’ risk weights and improve post-   stress capital ratios but that are not adequately supported or reflected in PPNR or loss   estimates

ii. “Flight-to-quality” assumptions and funding mix changes that increase deposits and   reduce the dollar cost of funding

iii. Significant  balance  sheet  shrinkage  with  no  consideration  of associated with reducing positions in periods of market stress

iv. Operating margin improvement.

• BHCs that make favorable assumptions should have sufficient evidence that they can be   reasonably assured in the assumed stress scenario. BHCs’ 𝑅𝑊𝐴 projections should be based on   corresponding projections of on-balance-sheet and off-balance-sheet exposures and their risk attributes and should be consistent with the severity of  the stress conditions under each scenario.
• For general credit-risk exposures, BHCs should project balances for material asset categories   with sufficient granularity to facilitate application of regulatory risk-weighting approaches   associated with different asset categories.
• For trading exposures, BHCs should translate changes in scenario variables into risk-parameter   estimates that drive 𝑅𝑊𝐴 calculations (e.g., the potential for 𝑅𝑊𝐴 per dollar of some trading   book positions to increase in periods of higher levels of general market volatility). Where
• 𝑅𝑊𝐴 projections are based on internal risk models, BHCs should not assume any 𝑅𝑊𝐴 reductions from potential data or model enhancements to 𝑅𝑊𝐴 calculation methodologies over the projection period.
• In all cases, BHCs should document any assumptions made as part of the balance sheet and
• 𝑅𝑊𝐴 projection process and perform independent reviews and validations of balance sheet and 𝑅𝑊𝐴 projection methodologies and resulting estimates