I worked 5 years at a large financial services firm with several $T in assets, and as such I have worked with Cobol Developers, though I have not worked with Co

I worked 5 years at a large financial services firm with several $T in assets, and as such I have worked with Cobol Developers, though I have not worked with Cobol much directly. They had it splits between Cobol developers on one hand and then the AWS/Java/Javascript developers on the other. There are a few issues with this idea. One is that there is some new Cobol code being written where necessary, but the big firms are trying to get rid of these people. The companies still with a lot of Cobol do want to get off of it, and while this is probably a 20-40 year project, this is not a growing industry, which I think is risky. There are a lot of cobol devs being trained up by the contracting services, which are used by the types of places that still use cobol, so I don't think salaries will ever compete with just being a developer in tech. There is a ton of bureaucracy in the orgs that use cobol, and especially around mainframes in those orgs, so that drives a lot of developers away. Since Cobol devs are often supporting 1-5 systems that they are the only ones that know how they work, and cobol mainframe jobs are typically batch processes, expect to be woken up pretty often in the middle of the night as you are supporting flaky 20 year old code that breaks often, the cobol devs I work with were all doing middle of the night work 1-2 weeks a month. Even if Cobol salary is great I don't think its worth it. The only point I could see it would be good is as a consultant that specializes in moving off of Cobol. I think a Consultant that could sell to these banks that you know how to migrate systems off of Cobol / JCL / DB2 / Sybase / Oracle -> Java / C# / Postgresql, Now that is where the money would be.