Operational risk
Operational risk is defined as “the risk of direct or indirect loss resulting from inediquate or failed internal processes, people and systems or from external events” (Basel Committee on Banking Supervision(2001), § 547). Both the unbundling of the minimum capital requirement into risk sensitive measures of credit and market risk, and the importance of operational risk warranted the explicit capital charge for operational risk. In contrast to credit and market risk, the measurement of operational risk is at an early stage of development. Developing a capital charge for operational risk is challenging both because of a lack of agreed methodology, and because of limited historical loss data.
In response to these challenges, the Basel Committee developed three approaches for measuring operational risk capital requirements:
1. The Basic indicator approach provides a simple way to determine a capital requirement, based on a percentage of gross income.
2. The Standardized approach assigns a capital charge for each of eight business lines based upon a fixed relation between average industry allocated economic capital and gross income for each business line.
3. Advanced Measurement Approach (AMA) provides flexibility for banks to use their own internal measurement approaches subject to meeting rigorous qualitative and quantitative standards. One of the key objectives of the AMA is to provide incentives for banks to develop improved measures of operational risk and to encourage operational loss data collection efforts to use as a basis for calibrating the regulatory capital requirements. This is spelled out in BCBS (2002).
If the broad regulatory goal is to make capital regulation more sensitive to the underlying risk, what measure of risk is the right one to target? Many of the risk measurement tools are variants of VaR, which was mentioned before and is defined as the loss to the portfolio due to an adverse market move that is only exceeded a% of the time. It is thus a scalar measure (a summary) of a very high dimensional problem. What standards should such a summary measure meet?
Artzner, Delbaen, Eber, and Heath (1997, 1999) lay out a set of criteria necessary for what they call a “coherent” measure of risk. They include homogeinity (larger positions bring greater risk), monotonicity (greater returns come with greater risk), sub-additivity (the risk of the sum cannot be greater than the sum of the risks) and the risk-free condition (if the entire portfolio is invested in the risk-free asset, it should be risk less). Importantly, VaR does not satisfy the sub-additivity condition, implying that a firm could concentrate all of its tail risks in one exposure in such a way that the risk borne by that exposure appears just beyond the overall portfolio VaR threshold. It becomes clear again that the informational loss from attempting to summarize the portfolio risk in a single number from the whole density could be substantial.
In this regard, the second pillar can be thought of as the main load-bearing column of the regulatory framework. In the words of Estrella (1998, p. 192) this pillar would allow regulators to reap “the benefits of informed supervision. Mechanical formulas may play a role in regulation, but they are in general incapable of providing a solution to the question of how much capital a bank should have”. In contrast, the supervisory would enable the supervisor to evaluate the inadequacy of an institution’s internal risk management and capital decision processes along a number of dimensions. Importantly, this pillar should be a flexible approach that allows for differences across institutions. Such differentiation is necessary to accommodate variations in business mix, risk profile, legal structure, and level of sophistication.
Comments
Leave a Reply
