Parental Income and College Enrollment in America

Tip of the hat to Adam Tooze for flagging a every interesting data set on the nexus between parental incomes and college enrollment from the Equality of Opportunity Project. Piketty shared an astonishing graph that shows that college enrollment rates are literally a linear function of parental income. Figure 1 reproduces that graph. All data displayed below is from this spreadsheet.

Figure 1. Enrollment rates by parental income.

The data set allows us to dig much deeper. Figure 2 recomputes the same graph as Figure 1 but for enrollment rates in 4-year colleges. The percentages drop rapidly. Whereas 80 percent of the 80-percenters’ kids are enrolled in some college; only 50 percent are enrolled in 4-year degree programs. This is very significant as most white collar jobs require a 4-year college degree.

Figure 2. Enrollment in 4-year colleges and parental income.

Figure 3 displays enrollment in public or private “selective” universities. Clearly, the authors’ definition of selective is not very selective. Presumably these are universities where one is not guaranteed to get in.

Figure 3. Enrollment in selective universities and parental income.

Figure 4 displays the same graph for “highly selective” schools. These can be more properly deemed to be selective. Only 10 percent of the 80-percenters’ kids get in here.

Figure 4. Enrollment in highly selective universities and parental income.

Figure 5 displays the same for “elite” universities. These are highly ranked schools; probably making the list of top 30 or 50.

Figure 5. Enrollment in elite universities and parental income.

Figure 6 displays the same graph for “Ivy Plus” schools. A potential list of these schools can be found here—although there is no guarantee that the mapping is one-to-one and onto. Chetty et al. define Ivy Plus as the Ivies plus Stanford, MIT, Chicago and Duke. 

Figure 5. Enrollment in Ivy Plus universities and parental income.

It is hard to compare these five graphs because they have very different scales. In order to compare them we log transform the data. Figure 6 displays them all together on a log scale. Here’s how to interpret it: exponential growth looks linear on the log scale; if it grows faster than linearly on this scale, it means that the raw series is super-exponential—it grows very, very rapidly (which here means access is much more unequal). So what this graph says then is that access to prestige schools is largely restricted to the upper class.

Figure 6. All together now.

Table 1 displays enrollment rates in elite schools for selected parental income percentiles. Less than 2 percent of the kids of 75-percenters get into elite schools compared to more than 8 percent of the 95-percenters and some 14 percent of the 98-percenters.

Table 1. Enrollment in elite schools by parental income.
Parental income percentile Percent in elite schools
25 0.64%
50 1.05%
75 1.85%
80 2.35%
90 4.48%
95 8.25%
98 13.96%

These numbers dramatically underscore how higher education functions as a mode of class reproduction in modern America. We are very far indeed from the land of opportunity.

3 thoughts on “Parental Income and College Enrollment in America

  1. Adam shows us proof of what I (and many others) would have predicted. Income inequality makes a big difference. It’s a vicious cycle that will continue on forever, until changes are made to stop it. I am interested to know if the same pattern holds true for a country like Germany where the tuition is free. Is the cost of the education at the best schools in the USA a determining factor? What importance does your parent(s) career play in determining where we go to school? And so on…….

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s