QIS 2: Operational Risk Data

4 May 2001

The second tranche of the Basel Committee on Banking Supervision's Quantitative Impact Study is attached. This tranche of the study is intended to allow the Committee to develop a calibration of the operational risk charge, and hence assess with greater accuracy the overall impact of the proposed New Basel Capital Accord, by collecting actual operational loss data. The data requested will also assist the Committee in calibrating the Basic Indicator and Standardised Approaches for assessing regulatory operational risk capital. A draft of the second tranche was circulated for comment during April, and the Committee is grateful to the numerous banks, insurance companies and supervisors who provided responses.

Some basic information on operational risk was requested in the first tranche of the QIS survey. Banks should not delay the submission of that information, which should be provided with the rest of the first tranche of QIS data by 1 June. The Committee is aware that the completion of the QIS within the deadline has created a burden for banks. For this reason it has staggered the two tranches of the survey, and the operational risk data requested in tranche 2 should be submitted to national supervisors by 1 August 2001.

The attached paper and spreadsheet set out a format for submitting loss data. This format is based on 8 standard business lines and 7 level 1 loss event categories (which in turn are subdivided into 21 level two loss event categories). Guidance on the composition of the business lines and loss event categories is provided. For working purposes the Committee has also set out a series of loss effect categories. These are intended to allow banks to assess fully the impact of operational loss events in their P&L attributions. The survey also requests information on the other components of the Internal Measurement Approach formula.

The survey is intended to allow the consolidation of industry wide data on a consistent and comparable basis. The Committee is aware that at present there is no industry standard loss classification and measurement system for operational risk and it is clear that even those banks with well established reporting systems will need to map their existing data to present it in the format requested. Furthermore, banks without loss databases are unlikely to be able to collect and present data in the timescale available for this exercise. Nevertheless, the Committee urges all banks to review the survey and consider whether they might establish mechanisms to collect such data in future. The Committee will be conducting further data exercises on operational risk next year, based on this survey format and hopes that it will lay the foundations for improved data capture and reporting by banks, and hence more accurate calibration of the operational risk charge going forward. The process of completing the survey should help banks prepare for the implementation of the operational risk element of the New Basel Capital Accord in 2004.

It is important that banks complete the survey using the definitions provided. Where banks' definitions do not correspond to those in the survey, they should map them to those provided in the attached document. Banks that submit results that do not attempt to follow the definitions and the mappings provided in the survey will have to be discarded from the sample. In addition, banks submitting loss data also need to provide the related information on exposure indicators and scaling so that the loss information to be aggregated with the results submitted by other institutions.

The Committee appreciates that this exercise is lengthy and detailed. It believes, however, that its proposals will have increased accuracy and risk sensitivity where they are based on a sound quantitative foundation. The data requested is of fundamental importance to the development of the operational risk charge and the calibration of the overall capital framework, and to the extent that more accurate and complete data is received then the need for buffers or adjustments for uncertainty is reduced.

All data that is received will be treated as confidential. The Committee intends to provide feedback to the industry on the results of the survey. This will be done on a basis that avoids any disclosure of individual bank data.