AMA's qualifying factors makes a bank’s risk assessments more forward-looking and reflective of the quality of control and operating environments. Directives imply that any Operational Risk Management (ORM) system aiming to qualify for AMA status, must be aware of, and be closely aligned with, the business strategies of the firm and the external factors that could impact its risk profile.Download an Insight
On 14th May 2008, Randall S Kroszner, Member of US Federal Reserve, while speaking at the Federal Reserve Bank of Boston AMA conference, acknowledged that there are strong linkages between the Advanced Measurement Approach (AMA) and supervisory expectations for sound management of operational risk. Talking about Operational Risks and AMA, he stressed that AMA has specific qualification requirements that not only brings about risk management improvements but also promotes an enterprise-wide risk capture and assessment across all business units of an organization. It builds upon the longtime best practices of banks to develop techniques for identifying, measuring, and attributing capital for operational risk.
Undoubtedly, this statement has drawn attention of management and stakeholders to the virtues of AMA in an enterprise. Could this provide the much needed impetus to AMA and bring its long-anticipated benefits to the forefront? Well, if Federal Reserve emphasizes on AMA, no bank or financial institution can afford to ignore it. Most organizations, today, realize that it is imperative to strengthen the soundness and stability of operational risk management (ORM) practice by employing AMA, in order to ensure that it does not lead to significant source of competitive inequity over rival banks & financial institutions.
Today, development of internal models and the implementation of the AMA is a significant challenge facing the industry. The first step, however, is to ensure that banks comply with the qualifying criteria of AMA.
Best Practices to Qualify for AMA
AMA’s qualifying factors makes a bank’s risk assessments more forward-looking and reflective of the quality of control and operating environments. Directives imply that any Operational Risk Management (ORM) tools / systems aiming to qualify for AMA status, must be aware of, and be closely aligned with, the business strategies of the firm and the external factors that could impact its risk profile. Few of best practices to qualify for AMA are:
Integration - By adopting an integrated operational risk framework, companies can ensure that all ORM initiatives are sustained and aligned with the corporate strategy elements & methodologies (RCSA & KRI) used to measure risk and operate as a natural system. It should also support evidential data (Loss Events) of the organization, which can be furnished to the model in a useable manner.
Evolved Scenario Analysis - Scenario analysis broadens the data set to understand and gauge the operational risk events on different plausible scenarios. It gives an opportunity to think outside the box and helps in making AMA futuristic. This is a key to organizational efficiency and calculating economic capital.
Strategic Risk & Capital Allocation - When designing an ORM structure, the bank’s overall risk scenario should serve as a guideline. This includes initiatives like laying down a hierarchical structure that leverages current risk processes, developing risk measurement models to assess regulatory and economic capital, and allocating economic capital vis-à-vis the actual risk confronted. Centralized aggregation of operational risk information collected via various self assessments across the organization, further, provides useful insight for the desired hierarchical structure. The implementation of these concepts allows risk to be handled consistently throughout the organization
Correlation & Dependency - ven though all the potential loss events don’t happen at the same time, under stress conditions like 9/11, uncorrelated risks become correlated. Within the capital model, the bank should look to identify and display the correlated and uncorrelated expression of exposure. This will allow dependence between specific risk event categories in the context of the business, product, channel and budget to be dimensioned.
Risk Adjusted Return on Capital - The bank should take on the initiative of Risk Adjusted Return on Capital and translate that to internal initiatives that allow the management to be more reactive to potential events. This should, however, be done in a balanced manner so that the executive decisions are supportive of a wider scope that accepts balanced benefit of cost.
Today, AMA is considered the most scientific method of the measurement of Operational risk in terms of continuum sophistication and risk sensitivity. The loss model approach leveraged by AMA is mostly used by the internationally active banks in developed economies. As put by one of risk experts, “The Actuarial loss model approach has become accepted by the industry as the generic AMA for the determination of operational risk regulatory capital for the new Basel II accord. The important step in the measurement of operational risk is the collection of relevant internal loss data for operational risk for a period of at least five years which is comprehensive, relevant and clearly linked to bank's current business activities and future vulnerabilities to risks.”
Loss Distribution Approach based AMA
Basel has exemplified the way to follow the process. Based on the loss experience of 89 banks obtained through the loss collection exercise in 2002 Basel calibrated the operational risk for average banks; constructing a loss distribution fitted to the set of losses of the 89 banks and estimated the loss at the 99.9% confidence level to be 15% of the gross income. It segmented the losses according to different standard lines of business and developed loss distribution for each line of business. In doing so, it determined that for average retail bank the operational risk is 12% of gross income, instead of 15% for the average bank that includes all the eight lines of businesses. For the average trading and sales it’s the 18% of gross income.
AMA has evolved swiftly as a tool for anticipating the losses-reflecting different assumptions about how current trends will unfold, how critical uncertainties will play out, and what new factors will come into play. It is now accepted that AMA paints pictures of possible future risks and explores the differing outcomes that might result, thereby empowering banks and financial institutions to predict future risks and losses better than ever before. Many industry gurus consider AMA as the most viable method to predict loss events, and hence strengthen overall ORM framework within an institution.
Realizing these virtues, banks are busy revamping their management structure in line with Basel II recommendations to give AMA greater priority. More analysis of loss scenarios are reinforcing vigilance in areas of in-house vulnerability, such as information technology or settlement procedures, and recording of transactions.
Operational Risk Rating Measurement Model -
Once AMA structure have been established by an organization, adequate procedures are designed and implemented to ensure execution of and compliance with its policies at business line level. This includes identification and assessment of operational risk, inherent in day-to-day processes of the bank. Operational risk, taken within each business unit, is measured and set within certain tolerance levels so that there are few surprises in the event of losses. This assures the top management that, in case of any untoward event, the bank has the capacity to absorb the losses because it has set aside sufficient capital. Furthermore, within those tolerances business management can decide to either increase the risk by removing expensive controls and adds less expensive controls, or decrease the risk by adding less expensive controls and removing more expensive capital
This is the vision and true value of operational risk management and AMA paves the way to realize that value. Key steps to achieve this vision,
- The AMA’s measurement model is tied directly to the risks that are identified in the RCSA. It allows business management to transparently see how the risk measure is affected by their decisions about the business environments.
- The identified operational risk is then rated, combining “qualitative” indicators (e.g. action plan follow up) and “quantitative” indicators (average score of internal control). This rating will be used in the economic capital allocation process so that well rated business lines or units are rewarded with a reduction of economic capital.
- Once the residual risk has been rated, next step involves simulation of an aggregate potential loss distribution for operational risks using actuarial approach. Drivers of simulation method are frequency and severity of loss events which are calculated through empirical loss data.
- After identification and rating of risk, the capital required is estimated on the basis of zone (high, medium, low) where the risk is placed.
- Once this is done the standard statistical technique including extreme value theory can be applied to determine the loss distribution for each set or rating.
Operational Risk Economic Capital - To obtain operational risk capital for the entire bank, an ideal ORM framework usually segregates the bank into different lines of businesses, rates them individually, obtains capital for each of them, and finally aggregates up the capital for each of these lines of business. This involves concepts like:
- Loss Data (Internal & External) - The operational risk loss data, derived from pooling of individual loss experience, provides banks with an invaluable insight into the past frequency of the events and their impact. In addition of the loss information, the associated rating of business and control environment at the time of loss is also collected. Combined together these evaluations help banks derive the loss distributions for each of the ratings or risk classes.
- Credibility Theory - In the wake of insufficient loss data, the internal loss data is used to modify the loss distribution obtained for the entire risk class, using credibility theory. If an entity does not have a reliable mean loss as there wasn’t sufficient data to calculate it, and the class to which it belongs calculates its mean loss with great confidence, then actual mean of the entity is considered to be the one that equals the mean loss of the class.
- Stress Testing - Although loss data Approach represents a good estimate of potential losses under normal market circumstances, it fails to capture 'one-off' or 'extreme stressful' events. For such situations, AMA puts forth a series of stress tests scenarios and sensitivity stress tests that shed light on the hypothetical behavior of banks portfolio and the impact on its financial results under extreme market movements. They may be based upon parallel movements in a number of risk elements or in one risk element, upon actual historical scenarios or upon plausible future shocks.
Business Benefits of AMA
As AMA matures and gains both the support and the confidence of management, it is becoming increasingly valuable to the business. Perceived initially to be mere risk measurement techniques, its elements can be leveraged and aligned with business performance management. To be successful, however, such alignment must be based on a clear vision of the potential benefits. Few of the benefits are discussed below:
Identified & Assessed Key Operational Risk Exposures -AMA enables an organization to identify, measure, monitor, and control its inherent risk exposures of the business at all levels. Elements like Loss Database, Control and Risk Self Assessment, and Key Risk Indicator play an important role; enabling the organization to evaluate the risk controls, based on the identified inherent risk, and to measure the residual risk which remains after the implementation of controls.
Evolved & Enabled Efficient Allocation of Operational Risk Capital - Within the spectrum of approaches in Basel II, the AMA gives banks the flexibility to build their own models, taking into account their own structure, correlation and diversification effects, insurance mitigation, and other factors. The goal of capital modeling is to estimate expected losses, which should be taken into account in product pricing, as well as to calculate the Value-at-Risk for operational risk (unexpected losses) that have to be buffered by economic and regulatory capital. Those figures could then be integrated into a bank steering concept. With streamlined risk management process, efficient allocation and utilization of operational risk capital can be ensured.
Consistent Operational Risk Management Information & Reporting Capabilities - AMA promotes the analytical and data management efforts of the business and the operational risk team to develop reporting protocols that serve both the individual business and the central management team. The “top-down” view of operational risk includes considering actual loss data, near misses, causes of loss and near miss, risk assessment, scenario data, and key risk indicators; and reporting aggregate losses and trends, risk-assessment results, key risk-indicator results, and economic and regulatory capital to all relevant parties. It also enables a “bottom-up” perspective that is directed to the needs of the business unit management. These reports are customized for the business unit to facilitate risk mitigation, process improvement, and cost containment.
Improved Risk Measurement & Management System - Application of an ORM framework supports a cultural shift to a risk-smart workforce and environment in the organization. It ensures that the capital charge is based in internal model for operational risks. In this way, operational risks sort of replicates the opportunity offered for credit risks and market risks to adopt internal, and more sophisticated models under certain constraints and after supervisory validation with the expectation that more sophisticated approaches could imply, on average, a lower capital charge than basic ones. This tradeoff is beneficial as it provides banks with the incentive to substantially improve risk measurement and management systems.
Incentive in promoting AMA is that the banks should come face to face with risks on a strategic basis and deal with them in a comprehensive rather than a "siloed," manner. In the last couple of years there have been steadily intensifying efforts to adopt the most promising AMA techniques and estimation models for operating risk, to define industry-wide benchmarks, and to promote reasonable standards. To most industry gurus, AMA process serves as a catalyst that leads to better risk management and has significant impact on risk profiles and loss reduction. The AMA program can greatly facilitate the development and usage of economic capital models, with the effect of deriving a risk-adjusted return on capital and capital allocation.
Technology is a key enabler for successful implementation of effective AMA systems and large banks with disparate operations in silos are seriously implementing automated, integrated solution, which can combine, re-engineer, and replace their back office systems. An effective solution that implements AMA strategies, methodologies and risk reporting functionality to identify, measure, monitor, control and mitigate operational risk. It ensures that the organization’s internal systems and controls are “credible and appropriate”, “well reasoned and well documented”, “transparent and accessible”, and are capable of being “validated” by internal and external auditors. Moreover, it provides capability to ensure that the risk management practices are embedded across the entire value chain.