The number one issues I have seen in the IT industry is lack of incentive, primarily in the form of salary, lack of respect, and lack of a C-level working on their behalf at the boardroom level. I know sysadmins who supported entire 200-250 person, 7 branch companies single-handedly, but get 40k, never had a budget, and got refused for hires and simple stuff like "hey, the cabling here is from the 1970's, we need a contractor to come in and recable the offices", and being told "no, you do it, at the same time you support the whole company, and no- you can't buy the TIA spec book."
It's no wonder companies are hemoragging good IT talent left and right.
If I could offer a single peice of advice to a company, it would be to create a CTO and CIO position if they don't exist, and get a good one who advocates on the behalf of the IT department. I see most of these issues as management failures first, not technical, so don't come crying about cobol to me.
I've been online for many many years and I STILL do not understand why people miss this simple idea.
Once you are making noticably more than you are spending, earning 2x and spending 2x means you're keeping 2x.
I set my price: AUD$850/day, with a TINY bit of wriggle room, and that's final. I know my worth (from a skills perspective and market/economical one.)
Edit: misread. GST is NOT included in the price; Super is my problem and comes out of the agreed daily rate.
My own consulting rate is higher than that, and there's way more people with my skill set. One of the big differences is I work with companies that value code as an asset. It sounds like he's working with Banks which view code as a cost of doing business.
And depending on the urgency, his rate goes up. He's retired, and works at most 2 or 3 months/year and lives a pretty normal life and lives in a very small house. His neighbors were pretty stunned when he bought himself a Tesla, his previous car was a Nissan Note...
He told me he was usually called in for projects where they needed specific expertise when writing Java to interface with existing COBOL, or rewriting COBOL parts with Java and adapt other COBOL code to inter-operate with that. Apparently most COBOL stuff now runs in the JVM - which tought me: in 30 years, Java will be the new COBOL.
I'll also throw in the counterpoint that you get taxed heavily on these earnings and you (usually) don't get benefits.
That's why a couple years ago when I had access to a zseries, and was hearing about how "desperate" the banks were I took a look. What I discovered are salaries that are some of the lowest in the IT industry and a cynical attitude towards hiring. AKA, banks would rather pay $40k a year for a "system operator" which is generally the ground floor for learning anything about a mainframe than hire someone with a comp-sci degree and teaching them the system while promoting them.
So, no thanks (you can fill in a less polite version), the banks can go to hell.