By Dave Glass, CEO and Co-Founder of Electrum
Banks in South Africa face a difficult decision over their reliance on an outdated language system.
Back in the 80s, I was very lucky to be exposed to early home computers – my astrophysicist dad was an early adopter. I can recall excitedly writing my first programmes on the famous Apple II home computer. After some frustrating attempts with the Pascal language, I discovered the Logo educational language, which was much more fun for an eight-year-old. Getting the daisy wheel printer to hammer out a sentence onto paper was a huge thrill.
Around the same time, in the financial services industry, the Common Business Oriented Language (COBOL) had become the operating language of financial services. Responsible for running most of the world’s banking systems, the brilliantly designed and easy-to-read, archetypal mainframe language still today underpins deposit accounts, check-clearing services, card networks, ATMs, mortgage servicing, loan ledgers and other services.
Programming languages have come a long way. Kids nowadays have so many more options (check out Scratch), and frankly, a much more entertaining and educational experience. So too in business, it’s not only the languages but the ecosystems around the languages that have radically improved – the developer tooling, utilities, pre-built components, cloud infrastructure, and even Artificial Intelligence assistants. That’s not to say you can’t run COBOL on cloud computing infrastructure, but do you really want to? Sixty-four years after it was developed, COBOL itself isn’t the fundamental problem, it’s the ecosystem around it that is no longer fit for purpose, and the world has moved on.
Finding COBOL programmers today can be challenging due to the language’s declining popularity and the ageing workforce familiar with it. Younger programmers are not exposed to COBOL as part of their training and as a result financial institutions often have to spend inordinate amounts of money on in-house training. One South African bank had to launch its own IT academy offering COBOL learnerships to train programmers to manage its core system. Alternatively, banks often hire retired programmers and pay them exorbitant fees to manage these legacy systems. A 2017 article by Reuters reported that in the US, experienced COBOL programmers were earning more than $100 an hour “to patch up glitches, rewrite coding manuals or make new systems work with old”. It might be cheaper to pay more for experience in the short term but what happens when the knowledge and experience are no longer around?
In my world of payments technology, I still see plenty of legacy payment systems that run on COBOL. In particular, many of the batch payment schemes were designed in the heyday of COBOL, were built in COBOL and have survived into the cloud computing era.
There are two main reasons why banks still cling to COBOL: cost and risk. Change is expensive and banks often baulk at the capital outlay for new system builds, and when they do embrace change, they risk their operations and their relationships with existing customers – who are of course very cautious in matters relating to their financial affairs.
In 2012 it cost the Commonwealth Bank of Australia (CBA) $750 million ($1 billion Australian) and five years to replace its ageing COBOL-based legacy core banking platform with a more modern system. It’s easy to understand why banks are still holding on to an archaic programming language. A recent article in PC Mag cited a report from the International Journal of Advanced Research in Science, Communication and Technology (IJARSCT), claiming that 43% of all banking systems globally are still using COBOL.
Change is however inevitable thanks to ever-increasing competitive pressure, as neo-banks and fintechs, enabled by new technologies and progressive regulations, are able to run cheaper, faster and with more agility. Without modern systems, banks risk stunting their growth potential as legacy systems hinder their ability to meet service demands. Upgrading COBOL systems will not be a simple task and will require substantial financial investment, and extensive time. Modernisation positions banks for progress – ready to entice new customers with unique offerings agile enough to meet their existing customers’ changing needs. In the long run, the cost of not modernising far outweighs the cost of change. Moving to newer technologies can help banks tap into a larger pool of developers with the skills and expertise needed to support and innovate their systems.
At Electrum, our Modernised Payments in South Africa report into the South African payments industry revealed that there is a clear trajectory towards the adoption of instant electronic payments. This transition not only reflects evolving consumer preferences but also presents promising opportunities for all stakeholders within the ecosystem.
As banking operations continue to grow in complexity and volume, cloud-based solutions built in modern languages offer greater scalability, allowing banks to handle larger volumes of transactions and data more efficiently. They also provide greater agility and flexibility, enabling banks to adapt quickly to changing market conditions, regulatory requirements, and customer demands, crucial in today’s fast-paced and competitive banking landscape. Better integration capabilities will allow banks to seamlessly integrate with other systems, platforms, and services creating a more connected and interoperable banking ecosystem.
Financial services often connote new systems with unnecessary risk. The reality, however, is that old, outdated systems carry just as much risk. If you are running the core global operations of the financial system on code no one writes or understands anymore, that’s a huge gamble. Don’t move… and you will be left behind.
The future of banking lies in embracing next-generation payments technology and the potential benefits of scalability, agility, cost-effectiveness, and innovation make it a compelling proposition for banks looking to modernise their technology infrastructure and stay competitive in the digital age.
