A new report from Meridien Research highlights a serious stumbling block for financial institutions working to comply with the proposed Basel II requirements. Joining the ranks of problems associated with market data as firms move toward straight-through processing, risk data is also proving to be quite a challenge as firms move to automate the internal-ratings based approach (IRB) to calculating credit risk capital and surrounding data for credit-risk assessment and management. Meridien's report entitled, "The Credit Risk Data Puzzle: Do You Have All the Pieces?" examines the risk-data dilemma and how firms are looking to solve the problem.
"The data hasn't really been focused on too much," says Peter Keppler, analyst at Meridien Research, "But it's going to be a big hump that institutions have to get over. The credit-risk data is all over the place within these institutions and has to be gathered from each product type or business line and is often in disparate formats." He adds that some areas within a firm may have data that is more usable such as already in databases or flat files, while other areas may still be dealing with the data in paper files or spreadsheets, which makes the automation even more difficult.
Keppler uses an example of moving good standing credits into the delinquency area, when the data on that credit comes from different areas and has to flow to another area. "But it seems like there aren't a lot of specific tools that are available for automating the process," says Keppler. "So this is going to initially be a big manual search, collect and organize mission."
How are firms approaching this undertaking? Keppler says that there is no one technology to solve the problem, instead financial institutions are going to have to examine their business lines and processes in order to come up with a firm-specific strategy to automate the data movement. "There won't be a one-size-fits-all solution," adds Keppler. "Every institution is going to have to put together the appropriate pieces for themselves. They are going to have to kind of cobble together a system and processes for collecting the data on an automated basis and cleansing it and then hopefully they will have a reasonably consistent data model that will work across the different asset classes."
The Meridien report further breaks down the IRB-data collection process into five stages: collection and aggregation, data modeling and storage, infrastructure building, automation and maintenance and enhancement. Keppler points out that this process will mainly be necessary for the more complex financial institutions that seek regulatory approval for using a more advanced IRB approach for their credit capital calculations.
But what is compelling these institutions to undertake such a challenge? Keppler says that the more complex institutions are being encouraged by regulators to implement the more advanced IRB approach for calculating capital allocation mainly because the size and complexity of asset classes and business lines continues to increase as banks are getting bigger and consolidation grows. "There's going to be a point where some sort of size or complexity will be passed and those institutions are going to have to step up the ladder," says Keppler. In addition, he says that there are moderate incentives in the form of slightly reduced capital requirements if institutions can prove they have a sophisticated ability to monitor and manage their risks.