Solutions to the COBOL Conundrum

The majority of life insurance core systems still run on COBOL, but the language is considered passé by up-and-coming IT professionals and the institutions that train them. What’s an insurance CIO to do?

“In 1997, the Gartner Group reported that 80 percent of the world’s business ran on COBOL, with over 200 billion lines of code in existence and with an estimated 5 billion lines of new code annually.” – Wikipedia

It has been stated that you cannot stop at a stoplight, use an ATM, or otherwise live out your day without executing COBOL code.  In fact, 70 percent of all business transactions today are executing COBOL, according to a recent article in eWeek.

The life insurance industry is certainly no different, with the majority of systems performing the administration of policies or handling agent commissions written in COBOL. For example, CSC’s Cyberlife, Vantage, Repetitive Payment System, and PerformancePlus are all written in COBOL and are widely used throughout the industry.

With thousands of jobs available for someone with COBOL skills across the country in a variety of industries, why don’t more universities teach COBOL? In my conversations with various college professors, most laugh at the idea of teaching this “uncool” language.  Some say that they do not believe students would take the class. Students, they believe, want to learn skills that will enable them to create mobile applications and video games, and are not interested in learning COBOL. Let’s face it, there are lots of jobs available for computer science graduates programming in cool languages too.

This poses a significant problem for the life insurance industry. As COBOL programmers retire, with fewer new programmers learning the language and a significant number of our core system written in COBOL, what are IS managers to do?

There are several ways to mitigate this problem:

  • Convert COBOL systems into systems written in “cool” languages. This is the most impractical solution.  Simpler systems have probably already been converted, leaving the most complex systems to convert. Projects to convert these systems cost millions and produce no business value, and in fact will probably result in negative value. Legacy COBOL systems have matured over decades.  Remember the old adage – hardware will eventually fail and software will eventually work.
  • Outsource the maintenance of your COBOL code to a vendor. For many this may be a viable solution. Others might ask – if they can maintain COBOL code, then why can’t most life insurers?
  • Encourage universities to offer the language – preferably as a requirement. As stated above this is problematic, but we can sit on advisory boards and perhaps offer grants to colleges to incent them to teach the language.
  • Train COBOL programmers.  Life insurers can do this themselves. This is certainly what the aforementioned vendors are doing. COBOL, given its English-like syntax, is not a difficult language to learn. And you don’t need to train computer science professionals. People have learned COBOL from all walks of life for decades. There are plenty of students that have majored in mathematics, biology or even music who are having difficulty finding work in today’s economy, and have the aptitude to learn COBOL.

J. Mark Clark // Mark Clark has over 31 years of experience in technology and the insurance industry. During that time he has worked as an entrepreneur, CIO, IT manager, consultant and software developer. He has vast experience running IT departments, managing large and small projects, building innovative applications, and installing and maintaining large systems. He holds an MBA from Michigan State University and is a devout Spartan fan.

Comment (1)

  1. Like the idea. COBOL should be encouraged in colleges. It is not difficult to learn though it may not be cool.

Leave a Comment