Does it really show that doing a bit better at the very top of the distribution is useful for something? I'd imagine it's a satisfaction type dynamic, where the

Does it really show that doing a bit better at the very top of the distribution is useful for something? I'd imagine it's a satisfaction type dynamic, where the benefits taper off.