Data Quality: The Truth Isn't Out There
REPORT PREVIOUSLY PUBLISHED BY OLIVER WYMAN
Financial firms that hold ambitions to improve risk management will need to be both resolute and creative to address issues associated with risk data quality and to achieve data mastery.
Since the crisis, there has been much discussion of risk measurement and management, but remarkably little of the data on which they depend. Poor data can undermine not only risk management but also most of the actions a bank takes. Be it a strategy decision or something as specific as setting prices for consumer loans, the story is the same: the quality of the decision depends on both the skill of the individuals involved and the information. If data quality is poor, the information will be poor, and only luck can prevent the decisions from being poor.
Paul Mee, coauthor of the paper and a Head of Oliver Wyman's Strategic IT and Operations practice, says, "Data is the lifeblood of any bank yet it is too often unattended, unclean, and unsuitable for the satisfactory functioning of the organs of the enterprise."
In the context of growing emphasis on risk systems and capabilities, data quality is an old theme that leaves unfulfilled expectations and disillusionment within virtually all firms. However, with a number of risk management renewal initiatives within tier 1 and even tier 2 institutions, there are rejuvenated efforts to address this aspect in a more concerted effort (again).
James Mackintosh, a Partner in Oliver Wyman's Finance and Risk practice, added, “Banks’ ability to develop and execute winning strategies is undermined by the quality of the fact base available to support decision making. Data quality is a significant strategic risk for many institutions.”
In this point of view paper, Data Quality: The Truth Isn’t Out There, we analyze the critical but often neglected discipline of management of information: information about exposures, risks, and customers. These all suffer from serious shortcomings. Poor data is routine, even humdrum; but, like an elephant in the room, poor data explains many of the problems banks experience, including insolvency.
What can be done? Data quality programmes start and often end with a framework consisting of measurement, governance, ownership, processes, and organisation. Although these are important, implementing such frameworks has failed to improve information at many banks. “Tried and tested” process-type data frameworks are failing the tests. We believe more creative approaches need to be considered.
This paper will examine perspectives pertaining to the underpinnings of data for effective risk management, examining how bad existing data issues actually are, the hallmarks of good data quality, and what can be done to improve it.