Financial institutions are under pressure to cut costs and keep up with consumer demands by replacing their ageing core legacy systems with modern solutions. But, as Parveen Bansal reports, they should look carefully at the alternatives before jumping.

The deafening mantra of ‘rip and replace’ is the cause of much misery for financial services institutions. These institutions are under pressure from solution vendors and consultants alike to replace their ageing legacy systems, which run the mission-critical applications of the business. Several reasons are given to back up the argument for changing but legacy systems still have their supporters.

The good news is that immediate replacement may not be necessary because solutions are available on the market that can enable organisations to best leverage their legacy systems.

Arguments for change

One argument for replacing legacy systems is based on the need to cut operating costs. As products become increasingly commoditised, “the strategic direction is to try to deliver the lowest cost per transaction, because cost-differentiation is the easiest thing to market”, says Stuart McGill, vice-president, EMEA, financial services and banking at Micro Focus.

With maintenance and support of systems said to account for about 80% of financial services institutions’ information technology budgets – compared with only 20% for new development – the pressure to replace old systems is even stronger. It is generally believed that by using modern technology, maintenance costs can be reduced significantly. This and the need for greater system flexibility are cited as two key reasons for change.

However, the value of the old Cobol-based legacy systems is sometimes underestimated and to replace the backbone of a bank’s systems is no easy task. “For 30 years the entire IT industry has been trying to migrate away from Cobol – unsuccessfully,” says Mr McGill. “The reason is that these Cobol systems work very well and so the value of migrating to new systems is not always clear.”

Built-in advantage

One reason for the prevalence of Cobol-based systems on mainframe and Unix servers is that much of the competitive advantage in terms of processing is built into the core systems. Thus the legacy systems hold incredible value for the organisation, representing the ‘family silver’.

David Pirrie, head of enterprise architecture coordination at Deutsche Bank, agrees: “Most legacy systems are not stand alone systems, they are integrated into the whole ecology of the bank,” he says.

Proponents of change often suggest that knowledge of the old Cobol systems is dying out. Mr McGill rejects this, however. “Cobol programming language is easy to understand and, with the continuous stream of Cobol-literate IT graduates from India, there is no danger of losing the ability to understand these systems,” he says.

In fact, Cobol remains the most widely used and accepted programming language in the world. According to industry estimates, more than 200 billion lines of Cobol code are used in today’s business applications, with several billion lines of new application code added annually. It is estimated that large enterprises have approximately $5000bn invested in these mission-critical systems.

“Cobol-based systems will stay in place as long as there is a business need for them,” says Mr Pirrie.

Risk to productivity

Mr McGill points out another possible problem with the move to new systems: “Current systems are highly optimised, a change to newer systems might actually result in a 20% or more reduction in productivity.”

And can any financial services institution afford even a small drop in productivity? The reputational, business and operational risks of such a project are so great that any change to the core must be considered carefully.

Mr Pirrie says: “The challenge for change comes about as a result of the continually increasing complexity of systems, as a result of much integration and interfacing between old and new systems.”

Large financial services institutions remain cautious and there have been no major implementations of modern vendor solutions to demonstrate that these alternatives to legacy systems are indeed scaleable, reliable and more cost-effective. James Beams, research director at Financial Insights, cites the lack of product complexity in these applications as one reason.

Scaleable and efficient

Legacy systems have not only worked well but, despite their age, they continue to work well today. They are highly scaleable, efficient and reliable for what they were built to do and in the way that they were built – to work independently of each other. So why change them?

Change is necessary to keep up with customers’ demands and the competition, and to adjust to the latest Basel II regulatory requirements to reduce operational risk.

Another bead in the string of arguments is that legacy systems are inflexible, making them difficult to change as customers demand more in today’s multi-channel environment. The Cobol systems were built at a time when the branch was the only point of contact between bank and customer. Today, however, there are a variety of customer access points, including telephone, internet, mobile phone and branches.

Today’s customers are demanding consistency of information across channels and for this the new systems must be able to communicate with the old systems so that they all offer a ‘single version of the truth’. With the arrival of the internet – where information is available immediately – customers have begun to expect a comparable service from financial service institutions.

Change of focus

Previously, business revolved around the product base of the organisation and applications were built for different products. Today, this is no longer the case. Competition has increased and customers are more aware of their needs; business has had to change its focus from products to customers.

The ability to integrate systems is now essential so that information can be exchanged between them. To keep ahead of the competition, financial services organisations must be able to make systems changes rapidly when they offer new products.

However, the good news is that change does not have to come in the form of core systems replacement ... yet. In the current tough economic climate it makes more sense to consider a preliminary approach, to renovate and leverage existing legacy assets while also taking advantage of new technologies. Thus organisations can buy time for a more risk-averse, phased approach to core systems replacement.

The move to a web services architecture or a services-oriented architecture offers banks an interim step before they embark on replacing the core Cobol systems.

Tools to reduce costs

With the complexity of the Cobol development environment in mind, Micro Focus offers organisations various integration and productivity boosting tools. One of these is Micro Focus Revolve, a suite of knowledge management tools that improves developer productivity and accuracy by accelerating understanding of the Cobol application and answering commonly asked questions. Micro Focus also offers tools to aid in the transfer of Cobol applications from mainframe to Windows or Unix platforms, thereby reducing costs.

The idea behind the web services model is to enable applications to communicate with each other, possibly over the internet, using a set of standards such as XML (extensible mark-up language), SOAP (simple object access protocol) and WSDL (web services description language).

The business benefit of using web services internally is that application development can be undertaken using a preferred programming language (Cobol, Java, C++) and architectures (J2EE, Microsoft .Net, etc). This can reduce the development costs that would be incurred if multiple interfaces had to be created.

Based on a foundation of standards, the idea of web services is to create a layer between front-end and back-end applications (currently on different platforms, written in different programming languages) and to eliminate the need for point-to-point integration and interfacing between various applications. This can be compared to a ring-road highway, which can be accessed using standard vehicles to deliver goods or information, regardless of the origin of the vehicles or the goods.

Thus web services can be used to increase the transparency of applications based on legacy systems. Micro Focus offers Cobol support for Microsoft’s .Net framework, as well as the legacy-to-web transformation and integration solution, EnterpriseLink – making access to web services easy for Cobol programmers.

Interim solution

Legacy systems may be expensive to run and difficult to change but they do still work. And there is no obvious, proven replacement system in the market.

However, the stress on legacy environments from customer expectations and the need to improve cost-income ratios continues unabated. Therefore, solutions that offer interim relief and allow organisations to continue servicing customers while reducing costs are to be welcomed, on the proviso that they are not a long-term solution.

Core system spending

celent predicts that large global institutions will spend approximately $3bn in 2003 and $6.5bn in 2005 on new core banking software, hardware and services. Global institutions with more than $100bn in assets spend an average of $1.2bn a year on information technologies. Half of this is estimated to go on core systems and their maintenance. When an institution of this size embarks on a core systems replacement project, it should expect to pay approximately $350m-$400m over the life of the project (typically three to four years). These costs can be broken down into hardware, software licence, system integration and software development. Return on investment will be slow in the first few years as the project is completed in phases. However, in a short time, the institution can expect to make savings from lower maintenance fees, the elimination of licence fees through consolidation of systems, easier deployment of new functionality, reduced operating expenses linked to staffing needs, and cheaper hardware through the elimination of mainframe systems.

Managing core system costs

Organisations that are replacing or modifying their core systems are beginning to take a more holistic approach to project management, rather than making changes in a patchwork fashion.

One of the primary tools that is being developed at major financial institutions is a more robust total cost of ownership (TCO) model.

TCO models provide a fact-based framework for financial services companies to manage their outsourcing options with vendors as well as to manage the challenging task of recovering centralised technology costs from their internal business user groups.

TCO models include all of the direct and indirect costs associated with maintaining a business application, as well as accounting methodologies on how to capitalise and allocate fixed costs.

Major components of these models are identified in the table below.

PLEASE ENTER YOUR DETAILS TO WATCH THIS VIDEO

All fields are mandatory

The Banker is a service from the Financial Times. The Financial Times Ltd takes your privacy seriously.

Choose how you want us to contact you.

Invites and Offers from The Banker

Receive exclusive personalised event invitations, carefully curated offers and promotions from The Banker



For more information about how we use your data, please refer to our privacy and cookie policies.

Terms and conditions

Join our community

The Banker on Twitter