Using Economic Capital to Determine Risk

Economic capital (EC) refers to the amount of risk capital that a bank estimates it will need in order to remain solvent at a given confidence level and time horizon. Regulatory capital (RC), on the other hand, reflects the amount of capital that a bank needs, given regulatory guidance and rules. This article will highlight how EC is measured, examine its relevance for banks, and compare economic and regulatory capital.

See also: How Protests Could Change Banking

Banks and financial institutions are faced with long-term future uncertainties that they intend to account for. It is in this context that the Basel Accords were created, aiming to enhance the risk management functions of banks and financial institutions. Basel II provides international directives on the regulatory minimum amount of capital that banks should hold against their risks, such as credit risk, market risk, operational risk, counterparty risk, pension risk, and others. Basel II also sets out regulatory guidance and rules for modeling regulatory capital and encourages firms to use EC models. EC as a concept and a risk measure is not a recent phenomenon but has rapidly become an important measure among banks and financial institutions.

(For background reading, see Basel II Accord To Guard Against Financial Shocks and How Basel I Affected Banks)

When banks calculate their RC requirement and eligible capital, they have to consider regulatory definitions, rules, and guidance. From a regulatory perspective, the minimum amount of capital is a part of a bank's eligible capital. Total eligible capital according to regulatory guidance under Basel II is provided by the following three tiers of capital:

• Tier 1 (core) capital: broadly includes elements such as common stock, qualifying preferred stock, and surplus and retained earnings.
• Tier 2 (supplementary) capital: includes elements such as general loan loss reserves, certain forms of preferred stock, term subordinated debt, perpetual debt, and other hybrid debt and equity instruments.
• Tier 3 capital: includes short-term subordinated debt and net trading book profits that have not been externally verified.

Note that these tiers may be constituted in various ways according to legal and accounting regimes in Bank for International Settlement (BIS) member countries. Additionally, the capital tiers differ in their ability to absorb losses; Tier 1 capital has the best abilities to absorb losses. It is necessary for a bank to calculate the bank's minimum capital requirement for credit, operational, market risk and other risks to establish how much Tier 1, Tier 2 and Tier 3 capital is available to support all risks.

EC is a measure of risk expressed in terms of capital. A bank may, for instance, wonder what level of capital is needed in order to remain solvent at a certain level of confidence and time horizon. In other words, EC may be considered as the amount of risk capital from the banks' perspective; therefore, it differs from RC requirement measures. Economic capital primarily aims to support business decisions, while RC aims to set minimum capital requirements against all risks in a bank under a range of regulatory rules and guidance.

So far, since economic capital is rather a bank-specific or internal measure of available capital, there is no common domestic or global definition of EC. Moreover, there are some elements that many banks have in common when defining EC. EC estimates can be covered by elements of Tier 1, Tier 2, Tier 3 or definitions used by rating agencies and/or other types of capital, such as planned earning, unrealized profit or an implicit government guarantee.

EC is highly relevant because it can provide key answers to specific business decisions or for evaluating the different business units of a bank. It also provides an instrument for comparing RC.

Performance Measure
A bank's management can use EC estimates to allocate capital across business streams, promoting those units that provide desirable profit per unit of risk. An example of performance measure that involves EC is a return on risk-adjusted capital (RORAC), risk-adjusted return on capital (RAROC) and economic value added (EVA). Figure 1 shows an example of a RORAC calculation and how it can be compared between the business units of a bank or financial institution.

Figure 1: RORAC of two business units during one year

Figure 1 shows that business unit 1 generates a higher return in EC terms (i.e. RORAC) compared to business Unit 2. Management would favor business Unit 1, which consumes less EC, but at the same time generates a higher return. This kind of assessment is more practical in a bottom-up approach. The bottom-up approach implies that the EC assessments are made for each business unit and then aggregated to an overall EC figure. By contrast, the top-down approach is more arbitrary, because EC is calibrated at a group level and then delivered to each business stream, where criteria for capital allocation can be vague.

Comparing to RC
Another use of EC is to compare it with RC requirement. Figure 2 provides an example of some risks that can be assessed by an EC framework and how it could be compared to RC requirement.

Figure 2: RC requirement and EC estimate

While a bank's EC figure is partly driven by its risk appetite (the desire for risk), RC requirement is driven by supervisory metrics set out in the regulatory guidance and rulebooks. Moreover, by contrast with regulatory capital models under Basel II, such as the advanced internal rating based (AIRB) model for credit risk, banks can make their own choices on how to model EC. For example, banks can choose the functional form and the parameter settings of their model. Therefore, EC modeling may adjust or ignore assumptions of AIRB for credit risk.

AIRB assumes that a loan portfolio is large and homogeneous, that longer-term assets are more risky, as reflected in the so-called maturity adjustment capped at five years, and that higher quality ratings have a higher correlation to reflect systemic risk. It also evaluates risk by rating classes and assumes a perfect correlation between rating classes and diversification within a rating class. (For more, see Measuring And Managing Investment Risk.)

Value-at-risk (VaR) models are typical EC frameworks for market, credit risk, and other risks. However, for credit risk, it is usually referred to as credit value-at-risk (CVaR). As an example, consider the loss distribution of a loan portfolio for relatively secure loans. The expected loss represents a loss that arises from the daily business, while the unexpected loss is the number of standard deviations away from the expected loss (the tail of the distribution). In the current example, suppose the unexpected loss is calibrated at the 99.95% confidence level, which corresponds to an 'AA' rating. Therefore, banks may calibrate their economic capital models according to management's risk appetite, which is usually in line with the bank's target rating. (For more, see An Introduction to Value at Risk.)

Some banks may use internally developed models to calculate their ECs. However, banks may also use commercial software to assist them in their EC calculations. A typical example of such software for credit risk is the Portfolio Manager by Moody's KMV, Strategic Analytics, Credit risk+ by Credit Suisse and CreditMetrics by JPMorgan.

EC is a measure of a bank's risk capital. It is not a recent concept, but it has rapidly become an important measure among banks and financial institutions. EC provides a useful supplementary instrument to RC for business-based decisions. Banks are increasingly using EC frameworks and it will most likely continue to grow in the future. The relevant question might be whether EC could one day supersede RC requirements.