June 2007 Capital Adequacy Extension © Copyright 2007, CCRO. All rights reserved. Page 41 of 92 available in measuring these types of operative risks, with their strengths and weaknesses. It is the responsibility of the CRO to determine what technique is appropriate to what risk. Figure 4.3 – Approach Spectrum to Measuring Operative Risks (Graphic provided courtesy of Ernst & Young, LLP) When assessing what measurement technique to utilize it is important to consider the following factors12: • Identify a meaningful set of operational risk indicators. • Determine the appropriate relationship among indicators (weightings and correlations). • Do your qualitative assessments translated into quantitative factors produce reasonable results? 12 Sound Practices for the Management and Supervision of Operational Risk. Basel Committee on Banking Supervision. Bank for International Settlements. December 2001. Qualitative Quantitative Hybrid • Experience and sound judgment of individuals in the company can assess and measure Operational Risk. • In some instances, the collective judgment of experienced individuals can “beat” the models. • Cannot always be relied on to systematically and consistently manage a wide range of risks that energy companies face in the market place. • Is an important component of the overall measurement approach but cannot be the sole source of measurement just as quantitative models cannot be one. • Most useful in instances where the hybrid and quantitative approaches are not capable of measuring Operational Risk in any meaningful way. • Mixture of judgment and quantitative techniques to address, lack of quantitative models and data. • Delphi based technique where collective experiences and judgments are collected and quantitatively ranked to indicate the level of exposure. • Where historical data or acceptable models do not exist, this approach can join the qualitative measure of risk with some quantitative indicators. • Develop operational standards (criteria to be measured against), gather data (data associated with complying with standard, loss-event data tracking, etc.), perform self-assessments (how current standards are performing compared to operational standards), and report through a Scorecard (convert qualitative data into quantitative data). • Combination of a severity model and a frequency model. • The severity model estimates the distribution of random variable to measure the central tendency of the occurrence. • The frequency model estimates the frequency of these events in the future. Qualitative Quantitative Hybrid • Experience and sound judgment of individuals in the company can assess and measure Operational Risk. • In some instances, the collective judgment of experienced individuals can “ beat ” the models. • Cannot always be relied on to systematically and consistently manage a wide range of risks that energy companies face in the market place. • Is an important component of the overall measurement approach but cannot be the sole source of measurement just as quantitative models cannot be one. • Most useful in instances where the hybrid and quantitative approaches are not capable of measuring Operational Risk in any meaningful way. • Mixture of judgment and quantitative techniques to address, lack of quantitative models and data. • Delphi based technique where collective experiences and judgments are collected and quantitatively ranked to indicate the level of exposure. • Where historical data or acceptable models do not exist, this approach can join the qualitative measure of risk with some quantitative indicators. • Develop operational standards (criteria to be measured against), gather data (data associated with complying with standard, loss - event data tracking, etc.), perform self - assessments (how current standards are performing compared to operational standards), and report through a Scorecard (convert qualitative data into quantitative data). • Combination of a severity model and a frequency model. • The severity model estimates the distribution of random variable to measure the central tendency of the occurrence. • The frequency model estimates the frequency of these events in the future.