From the Harvard admission website:
Admission to Harvard is need-blind, by which we mean that financial need is not an impediment to admission…
Financial aid at Harvard is entirely need–based and we are committed to meeting the demonstrated need of all students.
Other top US universities offer similar guarantees to prospective applicants. The full list can be viewed here.
On paper it’s a noble philosophy, a refreshing piece of evidence that success can be had without buying it. Looking down the road, policies like these at the gateways of our top universities shine light on the path toward class mobility and the un-disenfranchisement of our nation’s minority groups.
Sadly, according to the numbers it’s just false.
To show this, we first need to consider what makes a strong candidate for admission to a school like Harvard. If you ask anyone with professional experience in the college admission industry, you’ll likely hear him or her rattle off the following requisites:
Strong academic record.
Strong standardized test scores.
Compelling resume of extracurricular pursuits.
Powerful personal statement.
And herein lies the intrinsic problem with need-blind admissions: all four of these criteria are strongly correlated with the student’s family income
the private school problem
Again, from the Harvard admissions website:
We have worked hard for many years to learn about [secondary] schools in the U.S. and around the world.
In practice, this means Harvard knows which high schools are the good ones (it also means they know which schools have the students best able to pay full tuition: need-what?). They know that competition at private schools like Philips Academy is cutthroat. It’s not easy to crack the top 10% at private schools like Horace Mann or Exeter. Similarly, they know which schools are not so good, or ‘easy.’ And they keep this in mind while looking over your transcript.
The result is that these schools send students to ‘elite colleges’ at rates well above the norm. Here is the list of college matriculations from the 2012 class of Philips Academy. You will find none of the usual suspects missing from the list.
The apologist justification for this is simple. The kids at private schools are smarter! They were already vetted by a similar admissions process. So their curve is skewed! Of course their grades should be worth more. And beside that, they have better SAT scores.
But there is an easier explanation for the discrepancy in performance between private and public school students: family income. The likelihood of a private school enrollment increases with family income, and therefore these students are more affluent than their public school counterparts (despite the need-blind policies at schools like Philips Academy). 
According to the NCES (National Center for Education Statistics), academic performance (measured as GPA or # of As) improves as family income increases.  Mean SAT scores also increase as you ascend the income ladder. The discrepancy between students in the highest and lowest income brackets is nearly 400 points (out of a possible 2400)!  While these correlations may be explained by a number of factors, I believe the most culpable contributors are private one-to-one tutoring and reduced class sizes. SAT prep companies regularly boast of average score increases of 200+ points. It also helps that Chris Countryclub can afford to retake the SAT as many times as he finds necessary.
You would expect from the correlation above that a randomly selected group of high income students would sport higher marks on their report cards than a randomly selected group of lower income students. This is true in practice- A’s are handed out at higher rate in private schools than in public schools.  The improved performance in private schools is at best an indication of the students’ affluence and access to support services, and at worst evidence of systematic grade inflation.
Reduced class size has its own compound effect on the admissions process. 10% of US high school students attend private schools, but because of smaller class sizes those schools account for 28% of US secondary schools.  Therefore, there are more valedictorians and salutatorians per private school student than per public school student. More private school students (per capita) get these honors, because there are more honors to be had! What looks better on paper: salutatorian out of a class of 25 students, or ranked #80 out of a class of 1000? They are in the same percentile.
Admissions officers at elite colleges love to flaunt the number of valedictorians and salutatorians they reject each year. How many of those rejected valedictorians and salutatorians came from schools like Andover, do you think?
In other words, it’s statistically easier to attain academic achievement at private high schools! We should be adding gravity to transcripts generated by ‘bad’ high schools, not the good ones. But don’t say this to the wrong parent or you’ll have your tongue cut out.
So all the apologist’s exclamations really boil down to: the kids at private schools are richer!! To claim otherwise is to make the statement that your intelligence (or whatever word you’d like to substitute for raw academic ability) is correlated with your parents’ paychecks.
The NCES also tracks participation in extracurricular activities. Participation rates in every category of activity are positively correlated with income – except work.  In other, unsurprising words, an affluent student is more likely to join chess club, while a less affluent student is more likely to have a job. Who is the more compelling applicant on paper – the student who plays lacrosse, plays clarinet and borrows his parents’ car to volunteer as a lab tech at the nearest university on Saturdays, or the one who has only one entry on his/her list of extracurricular activities: bagging groceries for $7 an hour five days a week?
Research also suggests that low income students develop language skills at a slower rate than their higher income counterparts.  Meanwhile, access to a private tutor can drastically improve the quality of an admissions essay. Either legitimately, through careful criticism of successive drafts, or illegitimately. It’s pretty easy to hire someone to write your essay for you, and parents and students are willing to pay for a leg up. Check your local craigslist listings for more details.
Despite these advantages, admissions committees systematically give more credence to academic resumes generated by private and affluent public schools, which amounts to doubling up on class privilege. The proof is in the pudding – more than thirty percent of the typical Ivy League freshman class came from private high schools (based on a sample of 5 schools – Harvard, Penn and Columbia were not so forthright in their admissions statistics), compared to only ten percent of high school seniors. 
The proof is in the pudding and the result is fairy dust on the bottom line. At these schools, all of which promise to admit students NEED-BLIND and to additionally meet 100% of those students’ demonstrated need, nearly fifty percent of enrolling students do not qualify for any financial aid. Nearly fifty percent are paying full sticker price at a time when that sticker (~$55,000) is eighty nine percent the average income of families with school age children (~ $61,700) , including those families with more than one child who hopes to go to college! What a startling coincidence.
wait, hold on a minute…
I would like to paint a fair picture. A few things need to be made clear.
The discrepancy I am describing is more truly one between affluent and low-income schools, not necessarily between public and private. I narrowed my focus to private schools in the paragraphs above because they were an easier data set to manage.
I do not believe that these schools (affluent high schools and elite colleges alike) are maniacally bent on maintaining the status quo. The effects I described are unfortunate consequences of the system as it is structured, and are likely overlooked by admissions committees in willful ignorance. After all, if you take away all the assessment tools that favor affluent applicants, what’s left?
Meanwhile, these institutions have done many things right. Recently, the Ivy League has led a charge in substituting grants for loans in the financial aid packages of the lowest income students. Consequently, many of these students now graduate with their Ivy League degrees debt free. And the remarkable students enjoying this benefit deserve it- it is an incredible feat to get from where they were to where they are. Overall, students graduate from these schools with much lower debt than students at other private colleges- thanks to parental help on one end of the spectrum and better than average financial aid on the other.
Finally, the affluent children who benefit from these policies are absolutely qualified to attend the schools they do (minus the ones who cheated). I will vigorously argue the point that cognitive ability and socioeconomic status are NOT correlated. This means that there are smart rich kids, smart poor kids, and smart kids in between. There are more qualified applicants than spots in the Ivy League- Harvard could fill an entire freshman class with rejects from the previous year and still graduate a future Nobel laureate or two. These schools are not sacrificing the quality of their own academic communities by virtue of their admissions policies. Rather, they are collectively sacrificing quality of life on a much bigger scale.
the bottom line
We are placing an arbitrary limit on our progress.
Call me naïve, but I would like to think that our education system should aim to be that engine with which we improve our nation’s quality of life. It’s an engine with the potential to reduce violence and intolerance, drive our economy, and churn out solutions to all those problems challenging our species.
But for that engine to work properly, it needs to pull its raw material from all its available resources. Right now, the schools we consider our very best, those that are increasingly becoming portals to our nation’s most influential and lucrative positions, continue to draw fuel from the same old mines. Privilege continues to be held in the same hands that have always held it, and our universities are complicit in this aristocratic masquerade. They have not blinded themselves to their applicants’ need or their own bottom line. Instead, like a shameful creeper trying to hide his seedy gaze, they have just donned need-based sunglasses.
SOURCES AND RELATED INFORMATION
Table 158. Percentage distribution of elementary and secondary school children, by average grades and selected child and school characteristics: 1996, 2003, and 2007
Table 155. SAT mean scores and percentage distribution of college-bound seniors, by selected student characteristics: Selected years, 1995–96 through 2010–11
Table 165. Percentage of high school seniors who participate in various school-sponsored extracurricular activities, by selected student characteristics: 1992 and 2004
The claim that private school students have higher family incomes than public school students was deduced in the following manner, since the directly relevant statistics could not be found:
Rate of private school enrollment increases with income.
Therefore, private school student have a higher median family income than the general population of families with school age children.
Therefore, private school students have a higher median family income than public school students.
Current Population Survey by the US Census Bureau and the Bureau of Labor Statistics
Statistics were pulled from each university’s respective admissions website.