They are also much faster for many common loads.
[1] I suppose this is debatable. Oddly, many modern scripting languages use ancient compiler technology for no discernible reason. See also: the GIL.
Historical: most of them are 10~20 years old, "new compiler technologies" had not trickled down much when they were created and they had no real need for the things those provided (not to mention they have drawback, a GIL has better single-threaded performances than fine-locking which is important when single-treaded is your primary workload).
Also manpower, probably, it's harder to implement a generational concurrent GC than a refcounting or a basic M&S, and it's even harder to retrofit that into an existing codebase (hence pypy having pluggable GC backends and defaulting to a hybrid GC)
[1] Language developers sometimes claim they don't need it. I believe they're on the wrong side of history.
[2] Forks are common for this, and for good reason; often times languages that start with reference counting end up taking large dependencies on external libraries that manipulate reference counts (like Python with its Py_INCREF and Py_DECREF), at which point the language can't fully move to tracing GC without breaking its ecosystem.
Then again we may not get into this problem until there's a full GC which would be quite rare.
If fork() performance is important to you, you probably don't want compaction for the tenured generation at all. That's totally fine; while fragmentation will accumulate over time, it's no worse than C++. You'll still have copying for the nursery, but the nursery fills up and gets recycled so often that the COW optimization isn't really helping you there anyhow.