The World of COVID-19 Should Make Us Rethink Higher Education
Impact of COVID-19 on Higher Education
As the COVID-19 crisis continues to disrupt countries around the world. One industry that has been completely upended is education. Schools around the world have closed, impacting just about every student.
As of this week, UNESCO reported that 1.2 billion students have been affected by the Coronavirus, which represents almost 70% of all students. More than 160 countries have closed some or all their schools.
In the world of higher education, as colleges shut down and lose their revenue from student enrollment, it is likely many will permanently close. This will lead to a flight to quality. Schools with large endowments will be able to better attract capital to remain solvent, have more to invest (and have done so in the past prior to COVID-19) in digital platforms to continue services, and maintain connections with the largest companies that will continue to hire in this crisis.
It has therefore never been more important to consider the return to elite universities and who they benefit.
Do Elite Universities Create Opportunity?
We know that economic opportunity in America is very unequal today, but the degree of segregation by parental income across colleges is also very high. It is similar to the level of segregation across neighborhoods in the average American city
As an example the fraction of individuals in the upper income tiers who attend Harvard vs. the rest of the population is striking. Just over 70% of Harvard students come from the top 20% income bracket based on parent’s income (top quintile) and 15% come from just the top 1%. Most Harvard students are from the richest families.
Indeed, across elite colleges (Ivy’s plus similar elite schools like Duke and UChicago), high income individuals are vastly overrepresented relative to their fraction of the US population. Looking at the share of students who scored 1300 or better on the SAT (or ACT equivalent) by parental income quintile, we see that the fraction nationwide in the US is higher than the fraction at elite colleges except for the top income bracket. Those students with rich parents are overrepresented at elite schools.
However, we also know that children from low and high income families who attend the same college go on to have relatively similar levels of earnings in adulthood, suggesting that colleges can have an equalizing effect on intergenerational mobility, allowing low income individuals to rise up the income ladder.
Therefore, it seems obvious that elite colleges could serve as engines of economic opportunity if only they would admit more qualified, lower income students to help reduce economic inequality.
Reducing Inequality by Changing Admissions
Given that elite schools over index on students from rich families, it raises the question of how equalizing admissions could reduce inequality.
The equality of opportunity project ran several simulations to assess exactly this. First, they investigated how ignoring parental income might change the fraction of lower income students attending elite colleges.
To do this the researchers held fixed each school’s racial and geographic composition (since many students attend local schools). Within the population of all US students in each racial group and geographic combination at a school, they randomly sampled a student with the same SAT/ACT score as the student actually attending the school, repeating this process for all enrolled students at each school in their data set. They then match students to their new schools as a result of these random draws. This held fixed student ability as proxied by SAT/ACT score, but it ignored parental income as a factor in allocating students to colleges.
The researchers also tested another variant of this sampling where they added an offset to the SAT/ACT scores of lower income students since there is a strong relationship between test scores and income levels even after controlling for student ability. They calibrated this increase to the impact of being a legacy and/or an athlete in the admissions process, which is worth about +160 points on the SAT.
The results of these simulations are in the plot below. In blue are the fraction of students from the bottom income quintile that attend a range of college tiers. Less than 5% of these students attend Ivy plus schools, about 7.5% attend selective tiers, and 15% attend unselective tiers. In dark green and light green are the results of income neutral simulations with no SAT score boost for low income students and a +160 point SAT score boost respectively.
The income neutral selection process with no SAT score boost leads to less low income students at unselective schools and more at Ivy plus and selective tier schools; however, the increase at Ivy plus schools is small because lower income students tend to have lower test scores as noted above. Providing them a boost in the admissions process similar to what legacies and athletes already receive would lead to around a 300% increase in the number of low income students at these schools.
What impact does this have on inequality? One way to define inequality is the difference in the fraction of rich college attendees who achieve an income in the top quintile upon adulthood vs. the same rate for poor college attendees. If the fraction achieving this between rich and poor students is the same, then there is both low income persistence between parents and children and no relationship between parental income and selectivity of school attended (since attending a higher quality school does tend to lead to a higher income upon adulthood). Right now the rich-poor gap is 22%, meaning rich students are more likely to reach the top income quintile than poor students.
We can estimate how much an income neutral admissions process could reduce this gap by taking the results of the two simulations above and adding in the return to schooling, which the researchers estimated to be about 80%.
To add in this return to schooling, the researchers randomly assigned one of two potential incomes to students in each simulation. With 80% probability they assigned an income to each student based on a randomly drawn income from a student with the same parental income, SAT score to control for ability, and race at whatever new college they were matched to while with 20% probability they kept the student’s income the same as in reality.
They find that income neutral admissions policies could reduce the rich-poor gap and reduce inequality by about 15% in the case of no boost to SAT scores and reduce it 27% in the case of a slight SAT score boost.
The simple act of ignoring income in the college admissions process could dramatically cut the level of inequality in America by about a quarter.
Rethinking Higher Education
Given the high returns to tertiary education for individuals, but the clear bias in admissions at selective US institutions for children from richer families, nothing about the college process in America is meritocratic. The current state of affairs only worsens the rampant inequwality in US society, rather than ameliorating it, which should be the ultimate goal of educational opportunity.
With a huge shock to the world of higher education from COVID-19, forcing school closures and a reassessing of the value of attending college in a world where on campus services are shut down, there is a chance to remake the educational landscape towards a new vision that better promotes equal opportunity.
