At a New York Chapter meeting last spring, Martha Cummings, head of Risk, Banco Santander, identified data quality as a risk management gap that has always challenged the industry. She echoed the frustrations of many risk managers when she said, "The systems are never good enough. The data is flawed. And we don’t have enough resources to do what we need to do."

As you know, the regulators want us to overcome this challenge. They are demanding broader, deeper data sets and enhanced risk identification and control structures. In recent years, the industry and its regulators recognized that this need has become more urgent because of financial product innovation, globalization of capital markets, competition, and transparency. But our data collection and analysis objectives place an enormous strain on the legacy accounting and management information systems widely used in large, complex banking organizations.

To help our members improve data quality, RMA and Automated Financial Systems, Inc. (AFS) conduct an annual survey: the State of Enterprise Data to Support Credit Risk Management. Now in its fifth year, the study benchmarks the progress of large financial institutions in addressing challenges pertaining to data quality, integrity, availability, and transparency. Our large-bank members throughout the world will soon receive the 2012 survey. I urge them to complete it because this important issue can best be addressed as an industry effort.

One trend has emerged and is quite clear: Institutions are gradually improving the terms of how they gauge their bank’s quality of data—at least in the credit risk management area. In the 2010 survey, 46.5% of participants rated the quality of data within their institutions as either excellent or above average. Steady improvements have been made, with high quality ratings increasing from 27.5% in 2007 and 33.7% in 2008 to 43.2% in 2009 up to the current level. But this is how institutions rate themselves. I suspect that the regulators have a different point of view.

Nonetheless, one of the biggest challenges remains pulling data from disparate systems that are located throughout the bank—throughout the world, in some cases—and feeding one or a few data warehouses with information that’s current and accurate in real time.

Here are some key findings from the 2011 survey: