Existing models for studying the returns to college often present a single-dimensional account of the associated costs and benefits. Using such measures misses the web of interactions involved in the affordability of and returns to college. To take the clearest example, even calculating what an individual will pay for college involves capturing government, school, and private forms of financing, each of which carries different costs of capital to the student, distinct legal and regulatory constraints, and differing repayment structures. Similarly, imputing a basic benefit of college from the net present value of lifetime earnings against non-graduates, does not paint an accurately complex picture of the situation, for instance, a student’s perceived return to college will be dependent on how risk averse he/she is.
Moreover, it’s not only financial but also opportunity cost that is relevant for the decision of when, how, and whether to attend a post-secondary institution. In a generic situation, students forego four years of industry training and countless other avenues of personal and human capital development to attend school—a period of time that, for the majority, is in fact longer and less consistent than the formal duration of any given program.
A comprehensive model of the affordability of and returns to post-secondary education must account for the varied financing possibilities, risk profiles, and myriad financial and nonfinancial consequences of a college education. At the same time, there are differences in the access to information, and perception of financial information based on a student’s socioeconomic characteristics and cultural values. A student’s perception of debt, returns to college and affordability of college, along with their choices on schools and majors are influenced by where and how they get this information; from school counsellors, parents, peers and so on. We argue that an assessment of returns to education should also address, explicitly or implicitly, the impact of these more ‘qualitative’ associations.
What follows is our first pass at gathering a variety of papers that are representative of the frameworks that can be used to understand and model the true costs and consequences of post-secondary education. We have attempted to categorize the papers we summarize here based on the key themes they cover.
Debt and Student Debt
- Ambrose, Cordwell, and Ma (2015) study the effects of growing student debt on the financial flexibility of individuals to take on personal debt to finance small business startups. Using an ordinary least square regression model to capture unobserved differences among small businesses across counties, they find that an increase of one standard deviation in student debt reduced the formation of new businesses with one to four employees by 14% on average in each county between 2000 and 2010. Accumulating student debt reduces an individual’s ability to access different forms of credit, which in turn lowers an individual’s ability to start a new small business.
- Brown and Mazewski (2015) examine the effects of different types of debt on the intergenerational mobility of children from parents in the 25th percentile of the income distribution. Using the Federal Reserve Bank of New York’s Consumer Credit Panel dataset and Commuting Zone metrics (Q4 2000), they find statistically significant and positive associations between household debt and realized economic mobility of a child of parents in the lower income distribution.
- Dynarski (2014) argues that the student debt crisis in the US is actually a repayment crisis, where loans being paid when the borrowers’ earnings are the lowest and most vulnerable. As college enrollment has increased, the overall borrowing and the loans taken out by each borrower have gone up. The delinquency rate on student loans is higher than the default rate, and, moreover, that the outstanding amount of those who default is often less than the amount of those who are delinquent or repaying.
- A report from the Pew Research Center (2014) on key trends in the financial situation of a young adult household with and without student debt reveals that the net worth of graduates without student debt is seven times larger than graduates with student debt. In addition to this, students with debt are less happy with their financial situation and are less likely to see immediate returns to college education.
- A study by Dwyer, Hodson and McCloud (2014) on gendered effects of student borrowing shows that men and women have roughly equal access to debt, but differences emerge in the decision to take on debt, and in subsequent decisions about how to utilize the borrowing amounts.
- A report from Brown et. al. (2014) that uses Federal Reserve Bank of New York’s Consumer Credit Panel dataset from 2004-2012 attributes the growing student debt to higher college attendance, which adds to the number of borrowers and lowers student dropout rates. A sharp growth in the cost of college tuitions combined with readily available student loans exacerbated the debt problem. Between 2004 and 2012, the number of borrowers increased by 70% from 23 million borrowers to 39 million as well as the average debt per borrower also increased by 70%, from about \$15,000 to \$25,000.
- Debt accumulation has serious implications for social mobility and economic security of the middle class, according to this paper by Hyman and Ott (2013). In 2010, the consumer debt per person reached \$45000, which jeopardizes middle class wage growth and creates a situation of downward mobility. The authors attribute the rising debt to financialization of the US economy, which led to the decline of the manufacturing sector and stagnant wages.
- In this literature review, Dowd (2008) highlights several key factors and implications of student debt that are often not captured in causal studies on the effects of student debt, or are incorporated only partially. She calls for an interdisciplinary approach in understanding the implications of student debt.
- Using ethnographic and survey data to assess how parents and students get to know about financial aid, Tierney and Venegas (2006) suggest that being peer counselors places individuals in a fictive kin group that binds them together and confers a status on them as a college goer.
- Burdman (2005) investigates the debt dilemma for student who are averse to borrowing, especially their perception of debt and its impact on their opportunities. Her key findings suggest that low income student hesitate to apply for federal aid but take smaller loans and tend to attend low cost institutions.
- Barr (2002) argues for the removal of interest subsidies as a mechanism for improved access and quality for higher education. Instead he proposes a system where graduates pay an interest rate equal to the government’s cost of borrowing, a flexible fees system to assist some redistribution of teaching budget, and a wide ranging loan system with universal access.
- Using data from the National Longitudinal Study of Youth (1997) to assess the effects of indebtedness on college graduation, Dwyer, Hodson, and McCloud report in a 2012 study that taking out large loans and acquiring additional debt can reduces the likelihood of attaining a college degree.
- Fedaseyeu and Strohush (2018) build a model to analyze how access to federal student loans or government financing can lead to inefficient college entry. Their model is based in an economy with a two-factor production function with skilled and unskilled workers, and assumes that all workers are myopic. The authors indicate that inefficient college entry occurs when too many myopic worker obtain education because they overestimate the benefits of becoming skilled.
- Lobo and Burke-Smalley (2017) find that student loans produce higher returns for students compared to students who self-finance. Borrowing through student loans to finance education pushes the costs of financing into the future, thereby reducing the present value of such costs
- A survey by Wiswall and Zafar of undergraduate students at New York University (2016) reveals that students perceive a marriage market “return" to higher earning college majors and to completing a college degree. Furthermore, women in particular believe that science or business majors would raise their own earnings but also lead to lower rates and delay of marriage and reduce their expected number of children at age 30.
- Using the National Longitudinal Survey of Youth 1979 and 1997 waves (panel dataset), Webber (2016) uses a lifecycle earning simulation model to estimate the lifetime earnings premium of obtaining a given type of college degree. The author reports that all major categories have an expected return considerably greater than a high school degree. However, those lower in the ability distribution may face a more uncertain decision.
- Using data from the Population Survey (1970-2013), Abel and Deitz (2014) estimate the differences in earnings across students for three educational profiles. The authors estimate returns using a cost-benefit analysis model to report that students from all majors showed a greater return than seven percent even when the student is underemployed, indicating that college is a good investment.
- Using aggregate data at the national level and individual data from Washington State, this study from Long, Goldhaber, and Huntington-Klein (2014) finds that students’ choice of major responds positively to longitudinal changes in relative wages. Additionally, localized information on earnings may lead to stronger responses in major choice.
- Hershbein and Kearny (2014) make use of Census Bureau’s American Community Survey to provides estimates of annual and lifetime earnings for bachelor’s degree graduates of 80 majors. Looking at cumulative earnings over an entire career, the typical bachelor’s degree graduate worker earns \$1.19 million, which is twice what the typical high school graduate earns, and \$335,000 more than what the typical associate’s degree graduate earns.
- Zimmerman (2014) uses data from the Florida State University System to compare earnings of students who barely crossed the cut off to attend Florida International University against the earnings of the students who did not attend as they fell short. The financial returns to the marginal students of a year at a four-year college was about 11 percent.
- A report from Baum and Steele (2018) on financing of graduate and professional education shows that professional degree students heavily rely on borrowing. Professional degree students enrolled in public universities borrowed around $28,150, and professional degree students in private nonprofit universities borrowed around $36,670. On average, they borrow more than enough to cover their tuition and use loans to help cover their living expenses.
- As college education becomes more expensive, Baum and Ma (2014) call for a rethinking of college affordability as an investment rather than a consumption good. Understanding of college affordability requires an expansive approach that factors in prices, opportunity costs, returns to education, parental financial support, student debt, and variable educational outcomes.
- Based on interviews of students and parents on perceptions of Income Share Agreements (ISAs), this study from the American Institutes for Research (2016) found that parents strongly preferred ISAs to student loans. Students did not strongly prefer ISAs because of negative publicity and the fear of paying more than the borrowed amount under an ISA. Nonetheless, both parents and students desired flexibility to pay sooner and ability to predict amount of repayment.
- According to John (2006), greater emphasis on the roles of family income and financial assistance in decisions about preparation, high school dropout rates, and college enrollment, will help inform the debates about educational equality in ways that relate more directly to educational policy.
- A study from Calderone and McDonough (2006) that suggests lower-class families tend to spend the majority of their incomes on short term to immediate concerns over long term financial planning because of the limited and unstable nature of their finances.
- This College Board report (2006) argues that there can be no single percentage that answers the question of how much students can borrow without risking repayment difficulties. Instead, it recommends that borrowers making incomes 150% below poverty level should not make loan payments, and repayments should adjust to income and should not exceed 20% of total income.
- Dynarski (2002) reports a strong positive effect of introduction of scholarships on college attendance and yield rate of universities.
These papers show the promise of varied approaches to studying higher education and its implications. We believe that the next steps for research in education finance is to present a comprehensive picture of the the costs and consequences, both monetary and non-monetary, with accordingly adjusted returns. The papers in this literature review inform the kind of questions we ask ourselves as we began to work on the construction of an integrative model of the returns to higher education. With this iterative literature review, we hope to create a resource of the most pertinent findings on education from a range of disciplines to better inform the decisions of researchers, practitioners and policymakers.