Using that formula, how would management like to compensate Joe given the following:
* Steve is hired for $100k
* A year passes
* The company starts to lose money, let's say down to $1.5M a year.
* It is decided that Joe's industry knowledge - and only Joe's knowledge - had been making extra money for the company. It was some non-transferrable knowledge.
* Aside from the previous caveat, Steve and Joe had precisely the same skillset
* Joe is happily employed somewhere else for $150k
Time to a) can be short, particularly if the candidate is a good fit, but time to b) is typically three years. Early in my career, I took a little offense when a manager once told me that it would take three years to become useful. After some discussion, that wasn't a comment about my technical ability or level of work (they were actually very happy about that), but it was just a realistic timeline to get to b), the point that they could throw about anything at me and I could reliably pull resources and people together to get things done for the company in an independent and cost-effective manner. Depending on the work, time to (b) could be shorter or longer, but it's surprisingly long - longer than many want to admit to themselves. Over the years, I've looked at people in companies through this lens, and three years is typical.
Companies with an employee "replacement cost" model often have churn rates so employees stay with the company well below (b).
The "replacement cost" model also leads unwillingness to invest in employees, either by explicit training, or on the job training anticipating increased responsibilities. After all, they might be gone soon and we just have to go shopping for a replacement.
Longer term employees (typically the mid-level management) are then stuck because they're managing details that they really wouldn't otherwise need to, masking time to take care of problems that they would otherwise should be the sole owners of.
They're both subtle outgrowths of undervaluing people, leading to company underperformance.
It's still about minimizing expenditure to bring in a certain amount of profit.
That's how it doesn't work. Except if you sell bricks or something.
In the IT industry it's a surefire way to replace loyal and precious employess with talentless slobs, while you're thinking you're getting a better deal for your company.
Instead, it's associated with the best replacement that the company could find.
If Joe makes $2M a year for a company and gets paid $150k but Steve can do the same work at the same level for $100k, Joe's out and Steve's in. The initial $2M of value doesn't matter, except to lower the chance that a replacement could outperform you.
Cynical and unfair? Sure. That's how it works, though. Start your own company if you don't like it.