In 2017, the College Board made it easier for students with 504 plans to get the same accommodations on its exams that they received at school, including extended time.
A 504, for those not steeped in the argot of public schools, is not a handsome cut of Levi’s. It’s a document that describes supports that allow a student with a disability to participate fully in school.1
The College Board’s change was sensible and fair. Students still had to request approval, but they no longer had to demonstrate their need for accommodations from scratch. Requiring a new evaluation process was especially onerous for less privileged families.
Unfortunately, because we can’t have nice things, the situation went slightly wonky.
According to a recent paper from Princeton graduate student Tiffany Liu, more high school students began receiving 504s. Lots more. Particularly, wealthy kids. She calculates that the response to the new rule was 24 percent larger in the most affluent communities than in high-poverty schools.
This growing affluence gap in accommodations was already attracting notice. In 2019, Dana Goldstein and Jugal Patel from The New York Times reported on 504 plans in more than 11,000 high schools. In the richest 1 percent of school districts, 6 percent of high school students had a plan, more than double the national average. In Weston, Connecticut, where very few students are low-income, the figure was 18 percent, about eight times the rate in nearby Danbury, which is far less well-off.
There are two different jargon-y concepts to keep straight here. An Individualized Education Program, or IEP, is a detailed plan schools write for students who qualify for special education under federal law. It includes specific, annual goals and spells out the services the school will provide. The whole evaluation process is closely prescribed.
A 504 is different. When a student has an impairment that substantially limits “a major life activity” - including things like reading and concentrating - steps are taken to support the student’s full access to schooling. Accommodations might include extra time, breaks, etc. Unlike an IEP, a 504 plan does not require measurable annual goals or formal progress reports.2
Got the basics? Now, to the heart of the matter.
Affluence has become strongly predictive of whether a student will receive a 504 plan rather than an IEP. A 2025 study of Oregon schools found that 20 percent of students near the bottom of the income distribution received IEP services, compared with only 8 percent near the top.
For 504 plans, the pattern was the reverse. At the 95th income percentile, 2.9 percent had a 504. At the fifth percentile, 1.5 percent did. In high school, the gap was wider: 4.5 percent vs. 2.0 percent. In short, 504 plans were less common, but when they were written, they tended to be for wealthier students. Income differences were evident even among students attending the same school.
What’s going on? I spoke with the director of student services in a large district who explained that privileged parents often show up to her schools with a clinician’s letter already in hand. These families know how the process works and frequently have relationships with medical professionals to whom they can make their case. Families without that awareness and those connections may never seek accommodations.
Once students have an IEP or 504, income-based patterns continue. Researchers found that affluent IEP students were more likely to exit special education by high school - presumably because they no longer needed the services. When they received 504 plans, however, they were more likely to keep 504 plans in later grades.
Summing up, IEPs are much more common among poor children. 504 plans are more often commissioned for the wealthy.
Is this unfair? Not necessarily. There are benign explanations to consider. Lower-income students may have more profound needs that require different support in the form of full special education services. Poverty does horrible things to child health and development.
The rules have changed somewhat, too. Congress’s 2008 amendments to the Americans with Disabilities Act (ADA) broadened the definition of disability. In 2013, the diagnostic manual began allowing evidence of ADHD symptoms before age 12, rather than 7. That may have expanded the number of diagnoses.
Liu’s study raises another possibility, which is that privileged parents placed a higher priority on 504 plans as they became more valuable in the college admission scrum.
She compares high schools in states where all students took the SAT to others where they took the ACT. She also compares high schools with middle schools in SAT states. In each case, she concluded that when it became easier to receive SAT accommodations through a 504 plan, more students - specifically, high school students in affluent communities - got them.
I discussed this with two district administrators. They emphasized that this is about more than college entrance exams. They described something like a social contagion among families who hear about accommodations other students are receiving and then seek them for their own children. They said anxiety is the most commonly cited issue, especially amid intense pressure to earn high grades and passing scores on AP exams.
I was surprised by some of the accommodations being offered. Besides extra time on class tests, some students have the opportunity to test over multiple days, with access to the entire exam during each sitting. Some can take unlimited breaks while testing - even using that time for naps. In schools with cell phone bans, families are seeking 504s to allow their children to have their phone with them at all times, because being out of touch with their parents makes the children anxious.
The administrators feel unable to push back, even when they see accommodations as excessive or unnecessary, because families will press their case to senior district leaders or board members.
What’s Happening at Colleges?
While 504 plans have shifted the landscape of K-12 schools, only about six percent of high school students have them today. The transformation has been even more dramatic at the college level - especially elite colleges.
How dramatic? At Harvard, the share of undergraduates receiving disability accommodations grew from about 3 percent in 2014 to 19 percent in 2024.
In The Atlantic, Rose Horowitch reported that more than one in five undergraduates at Brown were registered as having a disability. At Amherst, it was 34 percent. At Stanford, 38 percent.
To be clear, these figures count students registered with a disability office, not students actually receiving accommodations. A student might register just to keep options open.
But the numbers are still striking, and they extend beyond these schools. Below, I pulled disability rates from the Integrated Postsecondary Education Data System (IPEDS) for a group of 65 highly selective undergraduate colleges going back to 2008. At that time, many schools were below three percent, at which point the school doesn’t even need to provide the specific number. Then, I added the data for 2017, around when the College Board made its policy change, as well as the most recent available year, 2024.
What do we notice?
Rates are up broadly. In 2008, not a single college on the list had 20+ percent of its students registered with a disability. In 2017, three did. By 2024, 23 of them did.3
Numbers at some schools shifted almost overnight. Princeton reported fewer than 3 percent of its students had disabilities in 2008. In 2017, same deal. Then, between 2017 and 2024, the figure rose to 24 percent. Cornell went from 4 percent in 2008 to 6 percent in 2017 to 27 percent in 2024. What accounts for such rapid change? I don’t know.
Similar colleges have wildly different rates. Georgia Tech and Harvey Mudd both specialize in engineering. Why did the former have a 2024 disability rate of less than 3 percent while Harvey Mudd was 24 percent?
The military academies have consistently low rates. Army, Navy, and Air Force were below three percent in 2024 and have never reported a higher figure. Why?
Very high disability rates occurred almost exclusively at selective schools. Less selective colleges tended to have fewer students registered with disabilities, not more, which is counterintuitive in many ways. To save space, I didn’t include a big list here, but places like Georgia State, Iowa State, Central Florida, Memphis, Oklahoma State, and Michigan State are below 10 percent. There are virtually zero less selective colleges with disability rates above 30 percent.4
What Accommodations Are College Students Receiving?
Are we talking about extended time on exams? Yes, that’s one of the most common examples. But there’s a far broader array of possibilities:
Advance notice that a professor will call on a student during discussion6
Housing accommodations, including single rooms or emotional-support animals
Some of these accommodations plainly remove barriers to learning. Others raise sticky questions, running the risk of softening course rigor or giving some students priority over others for scarce benefits.
Example: Stanford recently barred students with disability-based housing priority from extending that benefit to roommates without disabilities. That does not mean the original rule was wrong. But the university probably underestimated how it would affect the overall housing process for all students.
These dilemmas seem inevitable at prestigious, competitive campuses where 20 to 40 percent of students are now registered with disability offices.
What should we do?
My intent is not to criticize families who seek 504 plans for their children or districts who offer accommodations. I have neither the expertise nor visibility to imply that any disability designations are wrong. A high rate could mean a school is doing a great job identifying real needs, and a low one could signal indifference or neglect.
But when the perceived benefits of accommodations increase and affluent communities respond like it’s a Black Friday sale, we need to discuss it.
Maybe we can do a better job of evaluating students whose families cannot buy private assessments or hire an advocate. Maybe there are useful supports we should just offer all students, regardless of disability status, like extended time. Maybe there are updates to laws or policies that would clarify the appropriate circumstances where accommodations are warranted.
What we cannot do is pretend nothing’s happening. The numbers are too steep. In an era when court rulings have required colleges to remove any consideration of race in admissions, there is a risk that the path to acceptance at elite schools becomes associated with maxing out what can be written into a 504. I actually heard that term from someone while working on this piece - “504-maxxing,” like the Internet trend.
Accommodations should provide educational supports, not advantages. If we allow them to be stretched beyond their proper use and original intent, suspicion will grow. A backlash could make it harder for students who most need support to get it.
There’s a scene in Jerry Maguire where Renée Zellweger’s character and her adorable son are flying coach when the curtains between their section and first class are yanked closed. Her son asks what’s wrong.
“First class is what’s wrong,” she tells him. “It used to be a better meal. Now it’s a better life.”
Let’s not allow 504 plans to become the first-class cabin of schools.
504 plans get their name from Section 504 of the Rehabilitation Act of 1973.
Here is a slightly more detailed comparison of IEPs and 504s.
Only three schools out of 65 - Wellesley, Carleton, and Boston College - reported lower rates in 2024 than 2017.
The only two I saw were Pace University and SUNY Plattsburgh, both in New York. Anyone have a guess why? I poked around and couldn’t come up with anything.
Some schools provide the device for students. It might be a laptop that does not connect to the Internet. In other cases, students provide their own devices.
If you can’t find it, look under Classroom Access>No Cold-Calls.