Quite aside from the question of whether or not its okay to discriminate as long as it is principally whites who are harmed, I find this statement from the ACLU to be bizarre. I think it is fairly obviously wrong.
Moreover, the plaintiffs commissioned a study on the effect color blind admissions would have on the racial distribution of Harvard's undergraduate population. It should be taken with a grain of salt, as should the materials offered by the defense, and I suspect it overestimates the effect a bit, but it conforms fairly closely with anecdotal evidence I've seen and so sounds much more plausible to me than the ACLU's take.
It claims the share of black students would shrink from ~15% to ~0.9% and the share of hispanic students from ~15% to ~3%. Meanwhile, the share of white students would shrink slightly from ~37% to ~35%. The proportion of Asian students would roughly double, from ~25% to ~50%.
Again, the specific numbers might be a bit off, but it generally agrees with a lot of anecdotal evidence I've seen --- being black or Hispanic gives one a large advantage, being white is close to neutral, and being Asian is a large disadvantage.
If the primary goal is to be selected into Harvard, your finding is correct. Is it wrong to optimize for several metrics? What if the goal was to create opportunities that would exist for decades, for generations?
Moreover, the plaintiffs commissioned a study on the effect color blind admissions would have on the racial distribution of Harvard's undergraduate population. It should be taken with a grain of salt, as should the materials offered by the defense, and I suspect it overestimates the effect a bit, but it conforms fairly closely with anecdotal evidence I've seen and so sounds much more plausible to me than the ACLU's take.
It claims the share of black students would shrink from ~15% to ~0.9% and the share of hispanic students from ~15% to ~3%. Meanwhile, the share of white students would shrink slightly from ~37% to ~35%. The proportion of Asian students would roughly double, from ~25% to ~50%.
Again, the specific numbers might be a bit off, but it generally agrees with a lot of anecdotal evidence I've seen --- being black or Hispanic gives one a large advantage, being white is close to neutral, and being Asian is a large disadvantage.