Creating a Cutting-Edge Credit Reporting and Analysis Capability for Key Asian Banking Markets

Founded nearly 50 years ago as the Development Bank of Singapore, DBS was the catalyst to Singapore’s economic development during the nation’s early years of independence. Since then, DBS has become a very successful financial services institution.The bank’s strong capital position, as well as “AA-” and “Aa1” credit ratings (among the highest in the Asia-Pacific region), earned it Global Finance’s “Safest Bank in Asia” recognition for five consecutive years, from 2009 to 2013.

According to the company, DBS has a wide focus. In Singapore, it is a leading consumer bank and actively engaged in financing the growth of Singapore’s small and mediumsized enterprises, or SMEs. DBS even launched Singapore’s first real estate investment trust (REIT) in 2002.

Similarly, in Hong Kong, DBS is a leading player in the consumer and SME banking spaces, with leading market shares in equipment and trade finance as well as being a pioneer in the capital markets, with extensive product origination and risk management capabilities.

One of the key “back office” capabilities that supports many of these activities is credit reporting—the intelligence and skills needed to assess risk and creditworthiness for individuals and businesses. At the turn of the 21st century, this was still largely a manual and ad hoc process; one that was labor intensive, and as much art as science; and more importantly, it wasn’t working well. According to Sakthidaran Swamirajan, Vice President, Application Management at DBS, the existing process was built around a questionnaire and was not linked to data. For example, he explains, in the old process a relationship manager would typically ask an applicant a series of questions and the applicant could provide any answer they chose. The process was not linked directly to data.

According to the company, because the DBS credit determination process was very manual and it was error-prone, DBS assumed higher risk, and was required to hold more capital in reserve for potential bad loans.

To control costs, sustain growth, shield DBS from unnecessary risk, and grow the business, the company looked for ways to improve the existing process, strengthen its rules and build a more automated and fact-based process.

In 2009, the company’s credit process changed substantially. After briefly looking at alternative solutions, DBS chose to work with Progress, TIBCO, and Oracle to build a new credit determination process. “Progress Corticon gave us the ability to change the way the credit scoring is done in DBS,” says Sakthidaran.

The barriers to automating a complex process in a knowledge-rich environment are substantial. Many corporations like DBS have decision-intensive business processes with complex decision logic that must be applied at a high transaction rate. Thus, they are challenged with both getting the logic right and processing that logic fast enough to meet the required transaction throughput. And, whether decisions are made by people or by software, they must be made consistently and accurately every single time. With large sums of money involved, the stakes are particularly high, especially in a credit-reporting environment.

To master that challenge, in 2009 DBS adopted Progress Corticon Business Rules Management System (BRMS), with Enterprise Data Connector, as the cornerstone of its new system. DBS also employed TIBCO BusinessWorks and TIBCO iProcess Decisions along with an Oracle database as part of the solution. With eight scoring models, each with hundreds of rules, and hundreds of factors that go into a score, the system took several months for the DBS team to develop, but the results were worth the time spent, according to Sakthidaran. “With Progress Corticon we have been able to move at the speed of the market by implementing new rules to reflect new realities when needed. Corticon makes it easy to adopt and deploy these changes.”

Designed to serve internal customers, “Once set up by DBS, the Progress Corticon rules engine can read data from database tables, which is enabling us to change the key factors in the tables using online workflow modules,” explains Sakthidaran. “The rules that we use now are based directly on credit data and customer data. With our Corticon-based process we are capturing all the parameters. Furthermore, based on our data, we can change criteria on the fly; we don’t have to wait months to update the process,” he adds.

In the credit reporting system, the rules themselves come from a combination of regulatory sources such as the Monetary Authority of Singapore (MAS) and statistical analysis from the DBS Credit Portfolio Analytics Department.

Now that Corticon is powering the credit rules process, DBS routinely passes both internal and external audits by the MAS. This not only reduces their own credit risk, it also reduces the amount of funds which once had to be held in reserve to cover potential bad loans.

Like many organizations, DBS depends on many recurring decisions made correctly by professionals every day. The chance to capture, hone, and automate the decisions involved in credit reporting brought DBS to implement Progress Corticon Business Rules Management. “We are extremely pleased with the ease-of-use, end-to-end visibility of the rules, the rules completeness check, and the flexibility factors of mapping via tables, that we have been able to achieve with Corticon,” says Sakthidaran.

With the help of Corticon, DBS has built a better credit model; reducing risk and the amount of financial reserves that must be held aside to cover unanticipated losses. Over the longer term, Corticon is also helping to improve operational speed, delivering and updating new creditscoring models, reducing errors in models, and enhancing agility and adaptability. For example at DBS, Corticon helps support the reuse of logic across multiple models. And, above all, it has proven easy to use, deploy, and manage.

“With Corticon, IT and business can both respond quickly to market changes. Compared to previous practice, the turn-around time has been reduced by 25%,” notes Sakthidaran.