There are methods and practices that allow banks to assess risks effectively and to cover for the economic effects associated with risk-taking. These methods can be referred to as Economic Capital. In this chapter, the importance of understanding the relationship between overall economic and its building blocks is stressed. Moreover, making sure that risk assessment at an individual level is coherently and stylishly measured is of importance in this chapter.
Use of Economic Capital and Governance
Since the use of economic capital and performance has gone beyond relative risk measurement, the robustness of governance and economic capital, including control measures that surrounds the process, are very critical. Credibility and commitment from senior management are crucial elements for the capital process in a bank. Transparency in economic models is also another useful tool while making business decisions and undertaking risk management.
Among the many risk measures employed by the banks for economic capital purposes are: risk measure properties, type of product or risk that is being measured, availability of data, intended use of the risk measure, and trade-offs between complexity and usability of the measure. There are merits and limitations of each measure listed above which all needs to be clearly understood.
Practices used for measuring individual risk components are simpler as compared to the sophisticated technicalities in risk aggregation. With no theoretic consistency, risk aggregation techniques mainly rely on ad-hoc solutions and judgments. A limitation in technicalities of aggregation is validation. Despite assumptions being used, the overall risk may be undermined by the aggregation methodologies in effect.
To increase confidence in assumptions, validation is a good approach. Economical capital models can also be pointed out by validations, which is at a primary level. Areas like risk sensitivity can be assessed easily by economic models as compared to overall absolute accuracy.
Dependency Modeling in Credit Risk
Including linear and nonlinear connection between obligors, the modeling of dependency structures can be classified as one of the most crucial and challenging aspects in the credit risk modeling of a portfolio. Moreover, the Basel II risk weight function has dependency modeling connecting it with portfolio credit risk models. The correlation estimates’ accuracy is the biggest concern made in this section of economic capital and is still dependent on explicit and implicit model assumptions provided by current models.
Counterparty Credit Risk
The sophistication presented by measurement of counterparty credit risk ranges from gathering multiple system data, exposure measuring from millions of transactions ranging from as lately as overnight to possibly more than 30 years, netting agreement and collateral tracking, categorizing exposure across a wide variety of counterparties, and much more. The VaR model and Monte Carlo simulation approach are the most common modeling approaches employed by banks. However, these diagnostics are not fully comprehensive of all counterparty credit risk exposure despite being supplemented with complementary measurement processes.
Interest Rate Risk in the Banking Book
Simulation approaches and re-pricing schedules constitute the two main approaches for interest rate risk assessment. The calculation of economic capital by re-pricing schedule is made less suitable by the simple structure and restrictive assumptions. To determine the amount of economic capital, banks would use the magnitude of losses and their likelihood of occurrence, and the technique used will be dependent on the bank’s preference towards business type and earnings.
In order to utilize internal measures of risks not designed for regulatory purposes, these recommendations are poised to identify issues supervisors should consider:
- When assessing capital models, economic capital models should be used;
- Senior management should be committed to having a viable refinement of a bank’s economic capital;
- Transparency and integration into decision making;
- A robust process for identification of risk;
- Understanding the merits and limitations of all risk measures in use, within the context of their intended application;
- Risk aggregation;
- Credit risk dependency modeling;
- Counterparty credit risk; and
- Interest rate risk in the banking book.
Use of Economic Capital Measures and Governance
Economic capital will always be required by unexpected losses since in the process of accounting for a bank’s product and loan loss provisioning, expected losses will be accounted for. To analyze the economic capital, risks from certain activities or exposures should be identified, quantified, aggregated, and attributed or allocated some capital.
Management of credit portfolio, pricing based on risk, analysis of profitability of customers, and management incentives are some of the sections business line managers should perceive imposed by economic capital and offered opportunities.
Credit Portfolio Management
Processes of credit risk transfers are ways through which banks can assess risks or returns profile of credit portfolios and enhance their profitability. These activities are called credit portfolio management. Through this process, the creditworthiness of each borrower can be assessed in a portfolio setting.
Theoretically speaking, prices are exogenous to banks, thus acting as price-takers and assessing the expected return or performance of deals through risk-adjusted performance measures, assuming a competitive financial market. However, markets happen to be segmented in practice. The value-based management approach is typically incorporated in risk-based pricing. Shareholders’ value can be increased or decreased by economic capital through influencing the credit processes.
Customer and Product Profitability Analysis, Customer Segmentation and Portfolio Optimization
To measure performance at the customer level, the analysis of customer profitability by providing a broad and comprehensive cost view is paramount. This kind of analysis is useful in the identification of unprofitable customers, or those who are marginally profitable but use resources that need to be reallocated to more profitable endeavors. Through this process, economic capital thus becomes useful in the optimization of the risk-return trade-offs.
In a business unit management, influencing incentive structure directly affects the objective functions of decision makers by using economic capital. Most bank managers consider incentives a sensitive element in motivating their involvement in the economic capital allocation process.
Enterprise-Wide or Group-Level Use
Different business unit risks and different risk types are measured and controlled by a common currency provided for by economic capital. Market and operational risks are the main types of risks covered by banks’ models of economic capital. Generally, applying quantitative approaches to credit risk, market risk, interest rate risk in banking books and operational risk is useful. To mitigate contingency funding plans, banks view some risks as better covered.
Relative Performance Measurement
Economic capital measures play an important role in the computation of risk-adjusted performance for the purpose of assessing relative performance on the basis of risk-adjusted. Methods like risk-adjusted return on capital (RAROC), which is a relative measure, and shareholder value added (SVA), which is an absolute measure, are the most commonly used performance measure. However, setting a hurdle rate that reflects the cost of capital for the bank is a major limitation affecting the use of both RAROC and SVA. Compared to enterprise-wide capital adequacy assessment, the confidence level for the performance of business units is also be applied by some banks.
Capital Budgeting, Strategic Planning, Target Setting and Internal Reporting
Economic capital measures play an important role in the budgeting process. To facility business growth and operate within an overall risk appetite, most banks have established an internal reporting and monitoring framework. Banks have a number of ways of conducting capital planning. Most of them introduce forward-looking elements into the process. Common such methods include scenario analysis and sensitivity analysis.
What this means is that banks place more emphasis on qualitative rather than quantitative tools. They look at what would affect the bank individually or what would affect the market more broadly.
Bank uses the targets’ economic capital measure in corporate development activities such as merger and acquisition in conducting due diligence. The number of banks doing this is, however, smaller than those using this measure for other reasons described above.
Disclosure with authorities and dialogue with rating agencies are the major external communication channels where economic capital measures could be used. Most banks will disclose this information for each business unit.
Capital Adequacy Assessment
Banks have extended the use of economic capital beyond performance measures and decision-making processes; this is common especially in banks whose capital implementation is in the earlier stages. Large banks tend to use economic capital model for their ICAAP whereas smaller banks will most often use minimum regulatory capital requirements.
Large banks will use economic capital model for quantifiable risks while relying on more subjective approaches for less quantifiable risks.
Strong controls for making changes in risk measurement techniques is an important part of an effective economic capital framework. The viability of this will depend on commitment on the part of senior management.
The range of practice in regard to governance includes the following areas:
- Senior Management Involvement and Experience in the Economic Capital Process;
- Unit Involvement in the Economic Capital Process and Its Level of Knowledge;
- The Frequency of Economic Capital Measurement and Disclosure; and
- Policies Procedures and Approval Relating To Economic Capital Model Development, Validation, On-Going Maintenance and Ownership
Supervisory Concerns Relating to the Use of Economic Capital and Governance
Management needs to ensure that there are strong controls and governance surrounding the entire economic capital process. Supervisory concerns relating to the use of economic capital measures and governance includes:
- Standard for absolute versus relative measures of risk;
- Comprehensive capture of risk;
- Diversification assumptions;
- Assumptions about management actions;
- Role of stress testing including scenario analysis and sensitivity analysis;
- Definition of available capital where banks consider capital instruments that may be more loss-absorbing, more innovative, or uncertain forms of capital such as subordinated debt;
- Senior management commitment to the economic capital process; and
- Transparency and meaningfulness of economic capital measures.
The choice of risk measures has important implications for the assessment of risk. It is less clear how risk should be quantified. For this reason, the Internal Capital Adequacy Assessment Process (ICAAP) measures of capital under pillar 1 and pillar 2 should consider the impact of using different risk measure.
Desirable Characteristic of Risk Measures
An ideal risk measure should have the following characteristics:
- Intuitive: should align with meaningful notion such as unexpected losses;
- Stable: small changes should not produce large changes in the loss distribution;
- Easy to compute;
- Easy to understand; and
- Coherent: to satisfy the condition of monotonicity, positive homogeneity, translation invariance, and subadditivity.
Types of Risk Measures
Since there exists no ideal risk measure, many banks will often use a wide range of risk measures. The most commonly used are:
- Standard deviation;
- Value-at-risk (VaR), most used by banks for measuring absolute risk level;
- Expected shortfall (ES); and
- Spectral and distorted risk measure, allowing for different weights to be assigned to the quantities of loss distribution rather than assuming equal weights for all observations.
VaR and ES are the most commonly used risk measures. While the former is easily explained, it may not always satisfy the subadditivity condition, whereas the latter is coherent in making capital allocation and internal limit setting consistent.
Calculation of Risk Measures
In the calculation of risk measures, the following are often used:
- Confidence Level
Banks determine the appropriate confidence level for the economic capital model in their internal use of risk measures; this will depend on the bank’s target rating. Banks can, therefore, use a range of confidence levels for the same target rating with overlaps between different rating classes. The choice of risk measure and confidence level heavily influences relative capital allocation.
- Time Horizon
Risk measures depend on the time horizon used. An appropriate time horizon will depend on factors such as:
- The liquidity of the bank;
- Assets under consideration;
- Risk management needs;
- Regulatory requirements;
- Banks market standings; and
- Type of risk.
While market risk is estimated over a short period of time, such as one day or one week, credit risks are often measured using a one-year time horizon.
Risk measurement is typically performed at the portfolio level. For aggregation or decomposition, risk should be flexible and able to be computed at either a broad or a narrow level. Within a portfolio, it should be able to be decomposed to establish each subset, or be able to aggregate risk arising from several portfolios to convey a representation of the risk at the business unit level.
Supervisory Concerns Relating to Risk Measures
The stability in the computation of risk is important because the calculation of risk measures involves the use of simulation techniques. The degree to which economic capital is engaged in the decision-making process is affected by the availability of risk assessment at the senior management level that needs to be aware of differences between internal and regulatory measures of capital that originate from different risk measures.
Economic capital is calculated by first assessing the individual risk components, and then aggregating them to produce an overall economic capital measure for the entire bank. The process is based on identifying individual risk types and making the correct methodological choices.
Aggregation starts with the classification of risk types, combined to produce an overall economic capital measure. The classification may be divided into two types:
- The economic nature of the risk: market risk, business risk, operational risk, credit risk; and
- The organization structure of the bank: business lines or legal entities.
Range of Practice in the Choice of Risk Types
Many banks prefer to aggregate risk into groups by risk types across the entire bank before combining the groups. Other banks will use a mixed approach which combines both approaches, this is common where either particular business units or risk exposure are too small to be meaningfully measured separately.
Risk aggregation methodology used by banks has two components that is the choice of the unit and the approach taken to combine the risk. The way individual risks are combined relates closely to the scope of inter-risk diversification, that is the notion that combined portfolio would have a lower risk per unit of investment.
Typically Used Aggregation Methodologies
Banks will often differ in their choice of methodology for the aggregation of economic capital. The main approaches followed will include:
- Simple summation where individual capital components are added together;
- Applying a fixed diversification percentage, same to summation only that it subtracts a fixed percentage from the overall figure;
- Aggregation on the basis of a risk variance-covariance matrix;
- Copulas, which combine marginal distributions through a function; or
- The full modeling of common risk drivers across all portfolios, also called simulation.
Range of Practice in the Choice of Aggregation Methodology
There exist no established set of best practice concerning risk aggregation in the industry. Banks will often use a variety of approaches in setting values for the inter-risk variance-covariance matrix. These approaches include industry benchmarks, direct estimations, and expert judgments.
Supervisory Concerns Relating To Risk Aggregation
Meaningful aggregation of risk involves compromise and judgment to augment quantitative methods. Supervisory concerns with the economic capital aggregation relate to the validation of the inputs, methodologies used, and output of the process.
Validation of Internal Economic Capital Models
Validation is what provides evidence that an economic capital model works. It can permit a degree of confidence that the assumptions are appropriate, thus increasing the confidence of the end users.
What Validation Process Are in Use?
The more layers are provided, the more comfort the validation process is able to provide regarding the evidence for or against the model’s performance. Here is a list that moves from more qualitative to more quantitative validation processes
- Qualitative processes: involves use test, qualitative review, systems implementation, management oversight, data quality check and examination of assumptions; and
- Quantitative process: involves the validation of input/parameters, modeling replication, bench making, and hypothetical portfolio testing, backtesting, profit and loss attribution, and stress testing.
What Aspect of Models Does Validation Cover?
The validation steps outlined above can be used in assessing the most desirable properties of the models. The properties for which only week processes are available include: conceptual soundness, forward-looking and absolute risk quantification. Those that could be assessed using powerful tools include: integrity of implementation, risk sensitivity, relative quantification, and sensitivity to the external environment.
Supervisory Concerns Relating To Validation
Currently, there exists greater emphasis on the validation of models, with benchmarking of models being the main area of improvement. Users of the validation and senior management need to be informed clearly when validation is difficult or has limitations. In that case, they can be able to explore the potential costs of using a validation process that has not been fully validated.
Types of Models
Banks will typically use one of three credit models: Moody’s/KMV, Credit metrics, and credit risk+. Other banks may use models that are based on the asymptotic single-risk factor (ASRF), which is used for Basel II’s risk weights for credit risk.
Supervisory Concerns Relating To Currently Used Credit Portfolio Models
Since these estimates depend heavily on explicit or implicit model assumptions, supervisors can question the accuracy and robustness of correlation estimates used. This can significantly influence economic capital calculations.
When a bank uses a regulatory-type approach with single or multiple risk factors, the assumption may pose the following issues:
- Limited historical data on which the correlation is to be estimated; and
- The bank must account for concentration risk by other measures.
Counterparty Credit Risk Challenges
Since it involves gathering data from multiple systems, measuring counterparty risk can represent a complex exercise. The complexity of the processes indicates a need to have specialized processes and personnel.
Counterparty credit risk challenges may involve:
- Measuring exposure and measuring risk;
- Market-risk-related challenges to counterparty EAD estimation;
- Credit-risk-related challenges to PPD and LGD estimation;
- Interactions between market risk and credit risk wrong-way risk;
- Operational-risk-related challenges in managing counterparty credit risk;
- Differences in risk profiles between margined and non-margined counterparties; and
- Aggregation challenges.
Range of Practice
Institutions display a range of practices in measuring CCR in relation to the size and complexity of counterparty credit exposure. Most firms will adapt to either a standalone simulation engine or a value at risk CCR exposure engine which leverages well-developed and validated data and analytical system.
Counterparty Credit Risk Model Validation
CCR models for economic capital do not have specialized validation processes. However, they use the results of the validation process done by others to support the use of the model.
Interest Rate Risk in the Banking Book
Interest rate risk refers to the exposure of a bank’s financial condition to movements in the interest rates. Changes in interest rates affect an institution’s earing by altering interest-sensitive income, expenses, and the underlying value of an institution’s assets, liabilities, and off-balance sheet instruments.
The main sources of interest rate risk are:
- Repricing risk, which emerges when interest rates are settled on liabilities for periods which differ from those on offsetting assets;
- Yield curve risk stemming from asymmetric movements in rates; and
- Basis risk arising from imperfect correlation in the adjustment of rates earned and paid on financial instruments.
Interest Rate Measurement Techniques and Indicators
There are two basic techniques for assessing interest rate risk. They are:
- Repricing scheduling: which involves gap and duration analysis.
- Simulation approaches: which help banks in determining economic capital based on estimated losses (in worst case scenarios). It can be static or dynamic.
Main modeling issues in repricing scheduling may involve the type of simulation, the assumption surrounding the timing of interest rates, the holding period, and the time horizon.
For simulation approaches, the main modeling issues may include:
- Computational intensity derived from a large number of points along the term structure of the interest rates;
- A large amount of currency to track; and
- The availability of many related but different interest rates.
Main Challenges for the Measurement of Interest Rate Risk in the Banking Book
Here are some of the main challenges and pricing options for the banking book:
- Optionality in the banking book;
- Bank’s pricing behavior;
- The choice of stress scenarios;
- Scenarios based on historical distributions;
- Scenarios based on the principal component decomposition of the yield curve;
- Scenarios based on GARCH models;
- Scenarios based on options;
- Scenarios based on macroeconomic factors; and
- Scenarios linking credit and interest rate risk.
1) Copulas combine marginal probability distributions into a joint distribution. Which of the following is an advantage of copulas as a form of risk aggregation methodology?
- The effect of fixed diversification is sensitive to underlying interactions between the different components
- The method is easy to use as it estimates the inter-risk correlations and does not capture nonlinearities
- The simulation of common drivers provides outputs for calculating economic capital risk measures
- The method is more flexible than the covariance matrix method and allows for nonlinearities and higher order dependencies
The correct answer is D.
Through its flexibility in combining marginal probability distributions into joint distributions, copulas allow for nonlinearities and higher order dependencies, which the covariance matrix method does not allow for.