Does a College Education Provide a Positive ROI? It Depends

The cost of attending college continues to grow exponentially.  The how and why of these increases has seen lots of debate.   However, rarely does any writer try to refute the fundamental fact that college is expensive.  More often, proponents of college use this argument, "Sure college is expensive but look at it as an investment which has a tremendous ROI" (return on investment).  Actually, this is an excellent argument in the vast majority of instances. However, as I hope to outline in this post, the pay-off approach used is often biased. For example, a recent WSJ article pointed out the shortcomings of this argument for certain under-represented groups.  And this justification for the high cost of college does not fully consider all factors typically included in most business ROI models.

Whoa, you say, aren't you a marketer.  Are you getting ready to take us down a finance path?  The answer is yes.  Marketer or not, if the argument for a college education is based purely on the financial model, then some definitions are needed.   An informed customer needs some layperson terms and definitions to explain the ROI model.  It is also helpful to disclose total costs, investments, and periods associated with college and its subsequent benefits. 

Providing those definitions and then exploring if there is a positive ROI is the approach of this post.  I begin with a few definitions and explanations of concepts first.  After each definition, I discuss some problems and implications with these common definitions.  Then I provide some examples of the application to the college investment model or the ROI approach after making some common-sense adjustments to the current assumptions.  After reviewing these examples, we can see if we reach conclusions about the viability of college education on a purely financial basis.

I will begin with the most common cost discussed.  This is what is now becoming a federal government mandate cost, cost of attendance, or COA.  COA has its roots in the Federal Financial Aid literature, and this is a typical definition. "The COA includes tuition and fees; on-campus room and board (or a housing and food allowance for off-campus students); and allowances for books, supplies, transportation, loan fees, and, if applicable, dependent care. It can also include other expenses like an allowance for the rental or purchase of a personal computer, costs related to a disability, or costs for eligible study-abroad programs."

This is an acceptable definition to help compare costs across colleges and programs.  However, when used in the financial model for ROI (as it most certainly is in most cases), it violates a fundamental assumption of ROI models.  You should only include costs or expenses that would incur if you take the investment action.  That assumption would exclude on-campus room and board (or a housing and food allowance for off-campus students), transportation,  and dependent care.  The exclusion is based on the idea that a person going to college does not 'cause' housing, transportation, or the need for dependent care.  Even the need for a personal computer seems essential across all life situations these days.  However, the tuitions and fees are substantial and are a significant portion of the investment.  Associated with college would be the cost of student loans but only the portion of the loan related to tuition and fees.  Taking loans that allow for higher housing, transportation, or food standards is an issue for a different kind of blog.

An additional issue with the COA is that it excludes an essential cost, opportunity cost.  Opportunity costs in our ROI model would be forfeited income because most of the person's time is spent studying in class.  At a minimum, that would be 4 years of at least minimum wage in a full-time job, though it could be substantially more. Because it would seem likely that a recent high school graduate that earns admission to college is likely to secure more than a minimum wage job. 

Finally, the ROI model discounts future earnings and costs to a present-day value, so all dollar values are equal.  Any model should discount back to the decision point of choosing to attend college.  That would mean the first month of attendance.  The period is neither to a graduation date nor to the date of acceptance.  Therefore, with those assumptions in mind, we can explore some alternate scenarios. There is one other time-related assumption, the length of time to pay off loans.  Most federal models assume a ten-year time though research cited by US NEWS in 2014 shows an average closer to 24 years Link. Since most interest amortization is based on 10 years and most discount models in business use 10 years, we use the 10-year assumption. 

One other aside, we will not assume loan forgiveness as the current administration's policy will become standard practice.  If it should become policy, then the ROI would only increase exponentially.

Next, it is time to build out some scenarios.  This may seem tedious at first, but I assure you that I used the most unbiased sources I could find.  I also made sure to disclose the date of the studies. This is because college costs are changing, and estimates from even a decade ago can be poor representations of current conditions.  Oh, and by the way,  if bored with the numbers, you can jump down and see the conclusions.

First, we start by noting some of the research on the returns earned from a college education.  The first thing that jumps out is these returns really do depend on the degree earned.  This degree earned means not only what university is attended but also relevant is what major.  The recent WSJ article noting that under-represented groups were not earning a positive ROI neglected to account for this difference.

These factors will add quite a bit of variation in earnings.  We try to use on average with considerable caution that any individual should keep the college and major in mind.  One very commonly cited study comes from the US Census Bureau Link (however it is from 2002) showing over 10 years the college graduate earned $195,000 more than the high school graduate or $19,500 more per year.  Again not this study does not provide the complete story.   We have to factor in the opportunity cost of earning $75,600 over the four years of college, which we subtract from the $195,000.  Therefore, for this scenario, we will use $119,400 in additional earnings for the college graduate. The Pew Report from 2014 does not show a great deal of difference though the gap is decreasing Link.  For this scenario, the gain over 10 years is $175,000, and the opportunity cost is $112,000.  Therefore, we have an effective gain of $63,000.
​
Next, we need some estimates on the cost of tuition, fees, books, and supplies. The College Board Link  estimates in 2014-15 that the average tuition and fees for 4-year public institutions are $9,300 per year.  Books and supplies add approximately $700 for a nice round figure of $10,000 per year.  Assuming that the student borrows the 31% estimate provided by Sallie Mae in 2014 Link, and then the discounted amount of interest is added on. So we have another $13,500 of college costs.  Our total cost for attending college for four years is $53,500.

Now to revisit the additional earnings estimates for college graduates. Using the same discount rate on the cost of student loans, we have additional earnings of $103,725 from the 2002 census date versus the $53,500 cost of 4 years of college costs with interest on student loans. However, under the Pew Report of 2014, the picture is not so positive.  The discounted earnings surplus for the college degree is $54,730 versus the $53,500 cost of 4 years of college costs.

These examples are using averages.  Degrees in subjects that typically earn less (art history?) will see ROI go down and possibly be negative. On the other hand, the university's brand may raise earnings above typical or average earnings and thus increase ROI. However, using the more recent Pew numbers, should the amount of college funding by loans go up substantially and/or the cost of tuition and fees go up. The positive return on investment is likely to become negative.  While I am a firm believer in the additional non-financial benefits of a college education (see my previous posts), the ROI financial model with significant student loans just does not seem to make a strong case at this time.