Ability grouping within schools is criticized because high-SES children are over-represented in the top track and low-SES children in the bottom one. My point is that if you try to put all the children together to achieve equity, and if the high-ability learn little in the untracked classes, the high-ability children with affluent parents will leave. A static analysis will conclude that ability grouping increases segregation by SES, but ability grouping may lead to less segregation by SES than untracking when you account for the responses of parents.
This is true, and it's exactly what we did. My eldest sister did the same thing after trying a public kindergarten. Our local elementary schools are against acceleration and the principals I talked to over the years made that point crystal clear ("Why do you want a grade skip? Do you let your son play outside?").
My two youngest (6th/7th) are in private schools. My eldest (11th grade) attends a program run by the public schools. He attends it because it's a separate program at a community college that's run by two gifties. It's a wonderful program, and in part, the success is due to its distance from the other schools.
The public schools are presumably unaware of what they lose when they drive away affluent and/or intelligent parents. Or they don't care. I don't know.
