The companies I've worked at have been very bad at measuring things like "this is how much that bug cost us in terms of people who tried the app once and then never came back." They're so focused on the data sources that are easy to collect that things like this are basically invisible. They know churn rate is an important thing to track, but look at it much more from a "what about this new feature, or changing the color of the icons" perspective, than from a "what if our software was less of a dumpster fire?" perspective.
If you want to build quickly AND have flexibility AND have some reliability, you're gonna need some experienced employees, but if that's not a thing you've ever seen cause you've only ever worked with people fresh out of school... you're not gonna know what you're missing.
An older employee with experience can often see and avoid problems before they occur. A younger, inexperienced employee will not and will have to find the problem and attempt to fix it, usually three or four times before the problem is actually solved.
So, for six or twelve months, the employer can sail along gloriously, enjoying the benefits of cheaper employees. Until the bill comes due. And at that time no one can really help them except their current employees, who are the only ones who understand the history-dependent system they've built.
Fortunately, the time scales that most employers operate at rarely exceeds six to twelve months.
"A net gain" implies that savings in a youngster's salary exceeds the losses due to their inexperience. You're saying you know how all these add up in the neterprise?
I sure don't. I doubt most employers do either, since IT productivity is notoriously hard to quantify. Generally, it's more important to know what not to do. And that's something that only comes with time -- making mistakes and learning from them.
It seems clear to me that "even factoring in" means that there is a net gain for the employer, at a net loss to employees.