Salary-based higher ed stats may mislead

I have a passion for the quantitative. I major in math. I serve on the board of a quantitative nonprofit initiative. I review metrics for another. I get inappropriately excited when I solve a tricky statistics problem or merge two disagreeable datasets. I digress, but you get the idea: I love numbers. Yet I am also a student of the liberal arts, and I can recognize the limits of quantifying. Coming from this tradition, I stand behind President Craft’s criticisms of the Department of Education’s proposed higher education metrics. The Obama administration has suggested a variety of metrics to rate colleges. I want to object to one in particular: average graduate earnings. Assessing colleges based on the average salaries of their graduates is demonstrably misguided.

Let’s begin with the Department of Education’s justifications for considering this metric. In its synopsis of the President’s plan, the Department offers two justifications for this metric. First, that  “one purpose of college is to help students get a better, more secure career – and students deserve this information when choosing colleges;” and second, that “[f]or most families, choosing a college is one of the most important economic decisions they will make in their lifetime. Families should not have to struggle to figure out which colleges offer good value and which ones do not.” These arguments flow together: the consumers [students] need to know which firms [colleges] offer the best value [job]. This doesn’t hold water.

The conception of colleges as gatekeepers to salaries is preposterous. First, as noted over a decade ago by University of Chicago professor Andrew Abbot, “[a]ll serious studies show that while college-level factors like prestige and selectivity have some independent effect on later income, most variation in income happens within colleges—that is, between the graduates of a given college.” Abbot presented this point in 2002. Nine years later, the Census Bureau removed all doubt (shoutout to an expansive federal government). 2011 marked the first year the annual American Community Survey asked college graduates which field they earned their major in. The results are stark: what you study makes a big difference (consider subsequent analyses by The Hamilton Project and Georgetown University). By now, it’s clear that earnings gaps originate within colleges. Nevertheless, I believe even your major does not determine your income.

Realistically speaking, your income depends on two things: your profession and where you are working. This is not a revolutionary proposition – for example, Abbot made this same case in his speech. But it is still far too easy to overlook. Your major does not determine your profession: you do. Let’s consider an example: the highest paying major in the above studies is petroleum engineering. Why? Some professions, say, working in the oil industry, pay extraordinarily well (pun!). Thanks to this and a small sample size (very few students earn a petroleum engineering degree), petroleum engineering appears to be the best major to get. This ignores the fact that it is possible for an English major to work in petroleum engineering, if rather unlikely. This analysis extends to all industries: your major influences your income by means of affecting the profession you choose.

Next, let’s consider the impact where you live has on your income. As most of you know, cost of living varies. But the variations can be drastic. For example, most people consider a $60,000 salary terrific. However, if one earns this salary in New York City – a high cost of living area – it is only worth as much as a $26,320 salary in the Fargo area. It is clear to see that metropolitan colleges will have an artificially high raw income which may not mesh with their actual purchasing power.

The Obama Administration should not rank colleges based on graduates’ incomes. Publishing graduate earnings will unduly reward colleges that specialize in high paying industries and those centered in high cost of living areas. Neither where one studied nor what one studied determines his or her income. These variables merely confound the equation: firms determine income based on the industry and based on the cost of living. As President Craft noted, focusing on career outcomes is a narrow way to view college. I would like to add that this metric is also fundamentally flawed.


Leave a Reply

Your email address will not be published.

Welcome to the discussion. Before posting, please read our discussion guidelines.