Having said that, even if the academic outcomes of people with darker skin color were worse (no idea if that is true, or in which country) just knowing that is no proof. It may as well be cultural or even income based (pregnant mothers eating worse quality food for example, or same thing with young children).
Is the question even answerable? What kind of intelligence do we measure with "academic outcomes"? How do we adjust for income and quality of life as a child, things like having both parents, or grandparents? It's an absolute minefield of possible manipulation. Same data can probably be spun to "proove" any hypothesis you have.
Then imagine writing a proposal for such research. What benefits can you predict? If you find there is absolutely no correlation between skin color and academic outcomes after adjusting for all external factors will it convince people that are racist as part of their identity? I doubt it.
Lets say you do prove there is some correlation. What the result of that would be? Open discrimination and vindication of racist views. But even if the principle was true the average doesn't apply to individuals so you can't apply it to any particular person.
Intelligence is not just a single thing. We can assign a number by an intelligence test, of course, but there are lots of ways to look at it that are more nuanced. Generally, you're not going to say someone is better than everything you're going to find that someone is better at some things. And so can we direct people to the things you're better at for the better of everyone? Division of labor is a good thing.
But contrast, assuming that we find it's not true, then we can ask, well, why are we finding these differences in the data? if it's not genetic and perhaps we can find what it really is and then improve those groups that are currently disadvantaged.
The larger point is you have to know before you can make a useful adjustment.
This is especially clearly just a racist concept once you realize that none of the studies of IQ vs race actually do any genetic testing, they only rely on phenotype. Which is absurd given the fact that the child of a black-skinned person born in sub-Saharan Africa and a white-skinned person born in a village in Ireland is extremely likely to be black-skinned, but is genetically just as much part of both populations. It's only the history of racism that assigns this child as black more so than white.
And I would say that the racist claim is that there is observable correlation and that genetic explanations are about as probable as other explanations. But I suppose many non-genetic explanations (eg cultural differences) would also be counted as racism, since in these kinds of discussions race and culture are often jumbled together.