Erlang has a highly optimized concurrent GC as well. It's just optimized for different things. And maybe the concurrency of the GC is different; Erlang has one heap per process (aka green thread), and no concurrency within a heap.

Erlang GC is also very simple and easy to understand because language features only allow references in one direction. Much of JVM GC complexity would be wasted as there's no need to look for reference loops and such, since they're not possible.

It’s a GC tuned for imperative languages that prefer mutation over allocation, which is the exact inverse of what BEAM needs.