Toward a Better Approach to Calculating the ROI of College

There are an increasing number of approaches to calculate the return on investment of college. While most approaches include cost and earnings, how those variables are determined impacts the result, and how the result is presented to students impacts the influence it will have on their decision making.

Two approaches, one from the Center on Education and the Workforce (CEW) at Georgetown University, and the other from Payscale, allow future returns to be compared to an up-front investment today. The Brookings Institution takes a different approach, analyzing a college’s “value added,” which is the difference between actual earnings outcomes and the outcomes one would expect given a student’s characteristics and comparable colleges. A fourth approach, Third Way’s price-to-earnings premium, is used to determine the number of years it takes to pay off the cost of earning the degree.

While the existing approaches are helpful in getting a dialogue started and working toward designing an ideal approach for students, none are ready for game time. The limitations of existing approaches include, among other things, lack of personalization, the difficulty of interpreting the outputs, choice overload and not taking appropriate baseline costs and earnings into consideration.

Here are some considerations for what might be included in the ideal approach to calculating the ROI of college and how current approaches stack up.

Cost

Payscale uses the published “total cost” for tuition, fees, room and board, and books and supplies. They acknowledge a student’s actual net cost may vary. Brookings only looks at earnings, so doesn’t take cost into account at all. Third Way uses “total net price” using the average net price from the College Scorecard. Unfortunately, net price calculators on college’s websites aren’t always up-to-date or accurate, so it is difficult to find this information.

Incidental expenses should also be included, such as travel if a student is going to college far from home. However, a portion of room and board should be deducted as students that enter the workforce after high school without attending college still need to eat and live somewhere.

While most approaches take cost into account, none take debt into account. Two-thirds of students borrow to pay for college, and those interest payments should be included in the cost.

The ideal approach would use a close approximation of a student’s actual total net price, including incidental expenses, and expected interest payments on the debt they would be borrowing to pay for college, but would exclude living expenses they would otherwise be paying if they didn’t attend college.

Earnings

All approaches except CEW's take what a student would otherwise earn if they didn’t attend college into account. There are millions of high-paying jobs for high school graduates without a bachelor’s degree and, with many students asking the question whether they should attend college at all, it is important to take that opportunity cost into account. Payscale and Third Way compare earnings to a high school graduate without a bachelor’s degree, and Brookings compares colleges against similar colleges. Charles Clotfelter compares colleges against similar colleges in his book Unequal Colleges in the Age of Disparity, based on SAT scores.

All of these approaches use average earnings at the institutional level, whereas it would be better to use program-level earnings outcomes data. Although one-third of students change majors and many go to college undecided, for students who know what they plan to study, program-level earnings data would make estimates more precise, and such data have recently been made available.

Ideally pre- and postbachelor’s earnings would be compared within the same state or at an even more local level, and be based on occupations and/or intended major, as well as compared to earnings for colleges a student could otherwise be admitted to, to prevent undermatching.

Risk

All of these approaches make a major assumption (no pun intended): that students graduate from these colleges and thus get the benefit of the earnings associated with a bachelor’s degree. They don’t account for the risk of not graduating. Forty percent of students who start college don’t finish, and many of the students who do graduate take longer than four years. An ideal approach would incorporate the risk of not graduating and also include costs associated with a longer time period in school before earning postcollege wages. At the very least, approaches should include a very clear disclaimer that they assume graduation within x years, though students often have a bias that they will graduate on time, even when the data show they might not.

Time

Brookings and Third Way’s approaches evaluate ROI based on earnings 10 years after enrollment, while Payscale looks 20 years out. CEW’s approach evaluates ROI over 10-, 15-, 20-, 30- and 40-year periods. High school students are already suffering from choice overload, so having to evaluate and choose between multiple time periods could be overwhelming. The ideal approach would consider not only starting salary, but midcareer salary and lifetime benefits of a college education.

Time should also be taken into account for students considering alternatives to college. Fred Wilson, a venture capitalist, shows that the tuition at Flatiron School, a coding boot camp, can be paid back with six months of after-tax income for a high school graduate. This is based on starting salary, because boot camps emerged in the past decade and don’t yet have longitudinal data looking out 10 or 20 years or more. Furthermore, government data are not yet available for many of these programs.

There’s no way around the fact that calculating the ROI of college is complicated and involves a lot of variables and numbers. Students may have difficulty understanding the terminology, outputs and relevance of the various measures. An ideal approach needs to provide sufficient user education, which is critical for students interacting with financial information to make cost and earnings-informed decisions. At Edmit, the company I co-founded to help students make better-informed postsecondary choices, we’re working on putting this ideal approach into practice, and I welcome feedback on what readers think it should include.