I don't know. Well, it depends how you'd measure it. How would you measure improved excellence? The devil's in the details here. Are you going to get more publications, more higher-impact publications? I don't know.
Just as a logical argument, if your hiring initiatives exclude most of the population, then it seems that statistically, mathematically, you are not going to be hiring the best. The counter-argument would be that, somehow, in the normal hiring process, certain ethnicities are excluded through some vague mechanisms that are never clearly articulated. Just on a logical point, I don't understand how it can improve excellence.