At Marron, where Thomas Scott-Railton and Ilona Arnold-Berkovitz are presenting a paper on school integration.
Speakers are interposing between me and the slides, so pics aren't great. The resegregation of US school. 
Graph showing schools in Jersey. x-axis is B+H %, y-axis is free/reduced lunch %. Correl = 0.89, little stuff in middle.
Per Holme, US parents choose schools by high-status clientele (=no poor or nonwhite students), not curricular rigor.
Lon: school rating websites really just measure parental SES, not actual quality, reinforcing segregation.
The market incentivizes segregation through property value mechanisms; integration induces white flight.
The harms of segregation.
Thomas talks about the specific harms of double segregation, by both race and income.
US uni admissions incentivize segregation by rewarding families that send kids to elite secondary schools.
Affirmative action isn't enough because, to maintain donor bases, scholarship budgets, and rankings, it's limited.
Thomas: reward integrated schools.
E.g. give people who went to schools with high reduced-price lunch % a leg up in admissions.
Of note, it's not about more poor kids in uni but about giving the upper middle class incentive to send children to integrated schools.
Hartford magnet schools increased integration. Texas parents are willing to send kids to lower-performing kids for college admissions.
This is in context of Bush's class rank plan, where the top 10% of students in every school can get into UT Austin.
Caveat: it's important to avoid intra-school segregation.
There's undermatching: among students that show promise on the PSATs, 40% of whites and 30% of blacks take AP science.
Map of a Jersey suburb with a school map. Quality increases westward; so do house prices.
Under the free lunch bonus system, there would be incentives to move east.
Some US unis already take school SES into account, but don't publicize it well to parents.
This system is intended to be in tandem with both SES- and race-conscious affirmative action.
With the bonus for low-SES schools this could be displayed on real estate websites.
This still needs to be studied, with the exact metric to be optimized mid-implementation.
The inspiration for this is a This American Life interview with @nhannahjones from 2015.
Q from the audience: is it not a prisoner's dilemma since unis that don't implement the proposed policy piggyback on unis that do?
Thomas: this is already the case for affirmative action programs at unis today.
Lon: it's not about collaboration between unis so much as about marketing the school bonus points to real estate buyers.
Q: what about new studies from last summer indicating Republican parents no longer care about college?
Both this and the previous question are filibuster-length.
@fuller_brandon notes states with strong public unis can solve the coordination problem, like the Texas example. He suggests California.
Unlike the 2 previous people he did not filibuster.
@atboston asks how to avoid conflating race and class.
@atboston also asks about benefits to black and brown kids and about the dangers of reducing diversity to a benefit to rich whites.
Thomas: it's legality that pushes toward SES (Michigan's stricken-down plan gave a bonus to whites in black-majority schools).
Thomas brings up peer effects (e.g. from highly-resourced parents) as a benefit of integration to minorities and poor people.
Lon: the correlation between race and class in Jersey schools is so strong that SES affirmative action benefits blacks and Hispanics.
Q from me: Florida magnet schools have a lot of internal segregation at the class level. What happened in Hartford?
Thomas: we din't know but it's definitely important to avoid (of note, this was in the caveat slide).
Q: what about high-poverty schools that risk falling below a threshold for extra state money?
Lon: it's different in NY (where the Q mentions a 60% poverty cliff) but in Jersey it works more smoothly and w/less admin headaches.
Lon brings up a school that by marketing itself to rich whites as integrated went from 80% to sub-70% low-income.
Lon: said school kept its culture but now is being gutted due to top-down busing programs.
Q: why give bonus points to the poorest schools and not the most integrated?
Lon: we're not sure about the exact formula yet but the 90-100% low-income schools are the most difficult to integrate.
Both Lon and Thomas: giving the poorest schools the biggest leg-up also helps get their students into college.
Q: won't white parents, who choose schools mostly on race, keep finding ways to avoid majority-minority schools?
Q (cont'd): there are studies about the impact of introducing rich kids into poor schools and it's not always positive.
Q: won't this policy have the effect of inducing schools to recruit high-performing low-income students?
Q from @atboston: are there programs related to the bonus points actually giving resources to the lowest-performing schools?
Lon: at Rutgers there are programs to help students from low-income backgrounds acclimate abd succeed.
Thomas: the recruitment problem leading to brain drain is a serious concern but there are ways to balance it out.
Thomas: this is likeliest in general to integrate the middle, not the poorest schools.
Q: California has a 10% program for UC (though not the higher-end UCs); what was the impact there?
Q: without admissions transparency, how will parents be confident that they need to care about integration?
A (I forget whether Lon or Thomas): how much transparency is there about the importance of SATs in admissions? And yet people care.
