Have you seen any well-documented empirical evidence for this inverse correlation? I hear it often so I'm sure it has at least some merit, but I'd like to see some concrete data supporting it.