Is there any evidence there's a problem here worth solving (i.e. worth the non-zero costs of a change?)

Signs point to "no".