Not Need-Blind After All? "Top" Universities Agree to 9-Figure Settlement

I do see the tells differently than the schools literally seeing the family’s financial info and/or knowing their EFC during the admission process as some may in the examples I cited.

With that said, it’s no coincidence that year in and year out roughly 50% (give or take) of a given class is full pay at any number of need blind schools. Whether or not that truly makes a school need blind is beyond my pay grade.

6 Likes

They are likely need aware on a class level but need blind on an individual applicant level, because are are good enough at weighting the various correlates of need or lack thereof to get the desired level of need for the entire class despite some individual exceptions.

I recall the article that talked about Trinity College (I think…) and its admissions process. They had picked the class they wanted, but there was too much need in the pool so they had to start switching out needier kids for the less needy until they got everything to balance right. Of course, the school blind was need-aware vs. need-blind, but if schools are always coming up around the 50% mark with full-pay students, I strongly suspect that it’s not an accident.

It was Trinity College and here’s the article:

7 Likes

Not at some schools for some applicants like recruited athletes, Questbridge, Posse, and those from some college access orgs that a given school has a partnership with. These schools’ admissions peeps sometimes know the financial need/financials/EFCs of these applicants. That’s in addition to the waitlist students called out in the lawsuit.

I do wonder how much collusion was actually necessary. I mean I get wanting to use best practices. But I’m guessing that for relatively straightforward cases (looking at tax returns for W2 earners, ‘typical’ assets, # of kids in school), the results ought be pretty same-y from one college to the next, no?

I think this is essentially what many of these colleges meant by need blind. Like, they knew a legacy preference policy would, among other things, help them in terms of the net tuition budget at the end of the day. But that didn’t necessarily have anything directly to do with individual decisions, such that a high-need legacy could still benefit from the legacy policy. They just knew statistically what was likely to happen at the class level.

Where it gets interesting (to me, at least) is when they start using dynamic yield models to track how they are doing on various enrollment goals, including financial goals like hitting a certain net tuition budget.

If that sort of thing is affecting the individual decisions they make, calling that “need blind” is, to me, starting to be a real stretch.

To continue the example, suppose that their yield model is saying they are tracking too low on net tuition, but that if they start admitting more legacies, they will get back on track. So, they start admitting more legacies and get back on track.

That’s probably too crude of an example–although who knows. But I don’t know if I would personally call that sort of thing “need blind” any more.

Or if that doesn’t convince you, suppose we were talking about admits from independent private schools, or certain primary-residence zipcodes, or any combination of things that amounts to an increasingly accurate model of applicants likely to be low need?

How much of a dynamic “full pay by proxy” strategy can they have before “need blind” seems actively misleading?

And not to be cynical about it, but part of what is likely motivating these settlements pre-discovery is not wanting to have that sort of issue investigated and ultimately analyzed in full public view.

8 Likes

There are some well-known variables that are pretty material to some households. I mentioned before policies on work study and loans versus grants. Another potentially important variable is the treatment of net home equity. Another is using income thresholds for certain award levels. And so on.

Just to take a step back, the complaint alleges what they called the 568 Cartel came up with (and kept evolving together) something called the Consensus Approach, from which they derived the Consensus Methodology. A “methodology” is a recognized term of art for the formulas and practices that determine variables like that.

In fact, the College Board offers an off-the-shelf methodology known as the Institutional Methodology:

But CSS colleges do not have to follow the IM. For example, among other things, apparently the IM assesses parent net home equity at 100%, but individual colleges can do different formulas.

So the allegation is these colleges decided together the actual methodology they would adopt. It further alleges by doing so, they eliminated possible competition among each other to adopt more generous grant aid policies.

And I think factually, one thing that really hurt the defendants collectively is that some colleges started in the group, but then left it, and others refused to join.

Like, the complaint alleges Yale was an early member, but then claimed not to have participated from 2008 to 2018, and further claimed it did so because the Consensus Methodology was too restrictive in terms of the amount of financial aid Yale could offer.

To support that allegation, they cited this article:

The quotes in that article are, shall we say, not good for the defendants in this case. As are things like Harvard’s reported explanation for not joining (that adhering to the Consensus Methodology would have amounted to a decrease in awards at Harvard).

OK, so all that is pretty good circumstantial evidence the Consensus Methodology was in fact restricting competition in terms of generosity of financial aid. And that was very likely part of the motivation behind settling.

4 Likes

I found the article very confusing. It is clear that the colleges either deliberately engaged in shady practices or at very least, they broke the terms of the agreement that they had previously made with the justice department. So I am glad that they have been caught and forced to change (as well as pay damages), but I remain confused about what the impact of those practices were and how the practices benefited the colleges.

Most of the top comments on the NYT article all seem to be focused on anger about affirmative action (and sometimes DEI). There are also a bunch of comments about those colleges charging too much out of greed. However, the article itself doesn’t touch on any of those issues so it seems like the most of the commenters just want to vent about the things that they don’t like about those universities (including venting about a bunch of universities that were not part of the 568 group). All of that is fine, vent away, but I still don’t understand the impact of their illegal practices --I get that the colleges broke the terms of their exemption in that they were not truly need blind. What I am trying to understand is how the harm played out.

  • Is it that students who would have been admitted to the universities were not admitted to the college at all because of inability to pay full-tuition or they needed more aid? --I think maybe that is what the article is trying to say?

  • Is it that students were admitted but not given as much need-based aid as they would have been given without the group sharing financial aid practices? This seems harder to imagine simply because they are schools that provide quite generous need-based aid compared to most other universities. Having run net price calculators at many of them, the aid itself is quite good for middle and low income families. Even upper middle class families get aid from some of those schools.

  • On the other hand, my family’s experience is that despite the sharing of formulae, there is still a significant range in calculated aid packages. For example, University of Chicago, Vanderbilt, and Brown came out at nearly twice the net price as MIT and Yale was cheaper still when I ran their calculators for D22. So how did these practices actually help the colleges in crafting their aid packages?

  • Is it that the conspiring among the schools led to inflated tuition and fees that would otherwise be lower without this practice? Therefore it is not that aid wasn’t generous; it is that full-pay families were being charged more than if those practices didn’t exist?

  • All of the above?

You may have been typing your own post, but check out posts #26-27 (as well as #14) from @NiceUnparticularMan as he explains and hypothesizes in what seems a very plausible way to me, in a way that the NY Times article does not. It’s very nice having a generous attorney sharing his thoughts on the board! He chose an apt adjective in selecting his username!

3 Likes

In Yale’s case (as distinct from UPenn and others who were accused of considering ability to pay in wait list decisions) there really wasn’t much in the way of lies and opaqueness.

The plaintiff’s argument was that by giving an admissions preference to the children of large donors, Yale wasn’t truly need blind.

I’m not aware of anyone at Yale ever arguing that they didn’t give a preference to large donors, so… Lies? I don’t thing so.

Opaqueness? Well, I’m not aware of them publishing a price list to buy admission, so maybe.

As to not needing to care about financials, I don’t think any senior administrator at any major university is unaware of the importance of fundraising.

There’s a correlation here with endowment size and thus ability to meet need with scholarships/grants vs loans/workstudy. MIT and Yale have endowments double to quadruple the size of the other three institutions; Brown’s is smallest at ~7B.

1 Like

That’s the main allegation, that the formulas at these colleges ended up less generous than they would have been if the colleges had been competing freely against each other.

And they were in fact generous already, but that doesn’t mean they were as generous as they would have been without the agreement.

This sounds like what Trinity College does, and they are need aware. I do wonder if any schools that say they are need blind do this at the end of the admission round to make their net rev target? That would be over the line in my mind too.

Exactly. This is how some of these schools get to the 50% plus or minus full pays each and every year.

I agree. We will see if any of the remaining 8 stay in the fight. The fight that is getting more expensive every day, although a drop in the bucket for these schools. Regardless, a bad look for all involved.

2 Likes

To be fair, it is very hard to write a good general news article on a complex legal case like this, and I am interested myself, so happy to share what I have learned.

Incidentally, the complaint is here–these are just allegations, not proven, but it gives you an idea of what they are really arguing:

It is very long overall, but you can look up specific sections to see what they said about specific practices or colleges.

In terms of injury, that is on the bottom of Page 59 to top of Page 61, so just over a page. It is all very general, which is typical at this stage–they would need all sorts of data and then likely economic experts and such to estimate actual numbers.

Not something we have been discussing, but the next section, Relevant Market, is very interesting. Quick background, usually in an antitrust case you have to define and prove the relevant market in which you claim the cartel harmed competition. Here they define what they call the “Market for Elite, Private Universities” as “private national universities with an average ranking of 25 or higher in the U.S. News & World Report rankings from 2003 through 2021.”

Kind of amusing to see something like that show up in a legal document. I wonder if they would bring in witnesses from Reddit . . . .

They then go on to argue LACs are a distinct market, citing USNWR and the Carnegie Classifications, and also what they claim are lower admissions rates and higher yield rates at top universities versus top LACs. They similarly argue even public universities with high rankings are distinct, particularly for students needing aid in light of the OOS aid policies of these publics.

There is a lot more, but again it is very interesting to me to see these sorts of arguments mustered in the context of making out an antitrust claim.

2 Likes

Similar to our situation. S was admitted to two of the 5 schools listed and attended one of the others being sued. His need didn’t keep him from being accepted, and he greatly benefitted from the financial aid.

1 Like

Certainly motivates them (previously) to put most kids on the waitlist- they thought they could claim need blind while still looking at everyone’s need - outside of kids they outright accepted or rejected.

1 Like

These schools will be aiming to drop to 26 in no time! Plaintiffs hate this one simple trick!!

2 Likes

All those colleges that ended up tied at #24 must be miffed.

Edit: Actually, semi-seriously, the next private after that is NYU at #35. So that is probably going to help validate the plausibility of this market definition.

I do understand why their ability to meet need differs based on endowment and other factors. I just don’t understand how the practice of sharing formulas helped the colleges make their admissions decisions if in the end they offered such different packages. If for example, University of Chicago learns Yale’s formula, how is that knowledge helpful to Chicago given its endowment doesn’t allow it to offer as much aid as Yale? Plus, students tell the financial aid offices about other school’s offers when they file a financial aid appeal so what does having the formula shared tell them that is so helpful? Can’t they just look at the package, look at the students’ financials + appeal information, and basically work backwards to figure out the rough formula that the other colleges used? For that matter, wouldn’t running a net price calculator at other colleges give them a rough idea? However, I have not read the other documents posted here (about the “consensus methodology”) yet so maybe the usefulness of the formula sharing will be clearer after I do some reading.

So the allegation is these colleges not only shared information about their formulas, they actually agreed to use the same formulas.

Again, to get an idea of why the Plaintiffs alleged that, you can see this article they cited in which the Georgetown President seems to be saying exactly that:

That said, I do not know if these allegations are completely reconcilable with actual aid offers throughout this period. These things can get very complicated if there are agreements about some terms, other terms are left discretionary, there are changes over time, and so on.

2 Likes