What Happens to Students When the Federal Government Sanctions Colleges?

Header what happens to students 2 01
Academix Series Image
Part three in the Series

This report is based on research co-authored with Stephanie R. Cellini and Lesley J. Turner that is forthcoming in American Economic Journal: Economic Policy, “Where do students go when for-profit colleges lose federal aid?"

In July 2019, the US Department of Education (Department) officially repealed the “Gainful Employment” (GE) regulation that was put in place by the Obama Administration in 2014 to activate one of the Department’s most powerful accountability levers: taking away colleges’ eligibility to offer federal student aid to their students. The regulation restricted access to Pell Grants and student loans at for-profit institutions and career programs at private non-profit and public colleges where program graduates had high student loan payments compared to their post-graduation earnings. Proponents believed the GE standards would help protect vulnerable students from attending programs that left them with significant debt while yielding little economic value. But the current Department claimed that GE unfairly penalized for-profit colleges, and further, that penalties to schools in this sector would deny educational access to millions of students, many of whom most needed access to the economic benefits of postsecondary education.

At the root of the debate around regulations like GE is a key question: how do federal sanctions on institutions affect student enrollment decisions and the composition of the higher education market? Further, do these changes induced by regulatory policy improve student outcomes like student loan repayment? To answer these questions, this report synthesizes a new paper, in which my co-authors and I examine historical federal sanctions from the 1990s and 2000s that restricted students’ access to financial aid at some colleges and were enacted in response to concerns about rapid enrollment growth, poor student outcomes, and allegations of diploma mills within the for-profit sector. We map out a conceptual model of what happens to students when the for-profit colleges they attend or would consider attending lose access to federal student aid and use national data to estimate the impact of federal sanctions. These findings shed light on the tradeoffs that policymakers must face when balancing efforts to protect students and ensure access to high quality higher education—a conversation particularly relevant in light of the recent repeal of the GE rule and debates in Congress around reauthorizing the Higher Education Act (HEA).

The For-Profit College Landscape

Before jumping into our findings on how sanctions can affect colleges, it’s important to understand the landscape of the evolving for-profit higher education sector and the unique nature of for-profit colleges. For-profit institutions can legally distribute profits—the difference between revenue and expenses—to shareholders; this is distinct from public and private non-profit institutions that are not-for-profit, where profits are kept by the institution to fulfill its mission. As Figure 1 depicts, from 1995 to 2016, roughly three-quarters of all college-going students were enrolled at public institutions, with marginally more students attending four-year public colleges than two-year public colleges.1 Private non-profit students accounted for about 18-20% of total enrollment, with almost all of these students attending four-year schools. In the 1990s, only about 2-3% of students attended for-profit schools. But in the years leading up to GE, the for-profit college sector expanded substantially: enrollment more than tripled from 2001 to 2010, surpassing two million students—such that by 2010, about one out of every ten college students in the US was attending a for-profit institution. For-profit college enrollment has declined since the implementation of the GE rule, legal actions, and other regulatory and media attention, totaling about 1.4 million students (roughly 7% of students) in 2016.

A somewhat unique aspect of the for-profit college sector is that many for-profit four-year colleges also grant associate degrees and short-term certificates. Therefore, though enrollment in four-year for-profit colleges experienced the most drastic growth in the sector during this time, students earning credentials from for-profit colleges were more likely to earn associate degrees or certificates than bachelor’s degrees (see Figure 2). Moreover, among non-degree-granting colleges (schools that only award certificates rather than two- or four-year degrees), for-profit colleges enroll the largest number of students. From 1995 to 2016, about 200,000-400,000 students attended non-degree-granting for-profit colleges each year, as compared to 100,000-200,000 public college students and 20,000-30,000 students in the private non-profit sector (see Figure 3).

Access to federal student financial aid is an important revenue driver in the for-profit sector, with over half of for-profit institutions gaining 70 percent or more of their annual revenue from federal Title IV program funds (Title IV of the HEA authorizes the largest federal financial aid programs, including Pell Grants, work-study, and loan programs).2 Concurrent with the rapid enrollment growth, financial aid usage in the for-profit sector also grew substantially from 2000 to 2010: the for-profit college share of both Pell Grant and subsidized student loan disbursements more than doubled to about 25%, as shown in Figure 4. At its peak in 2011, the for-profit college sector accounted for about a tenth of the total undergraduate enrollment, but about a quarter of Pell Grant and federal student loan disbursements.

At its peak in 2011, the for-profit college sector accounted for about a tenth of the total undergraduate enrollment, but about a quarter of Pell Grant and federal student loan disbursements.

Figure 1: Fall enrollment in degree-granting postsecondary institutions

Image Alt Text Source: Table 303.25, Digest of Education Statistics 2017, National Center for Education Statistics. Note: Title IV institutions only.

Figure 2: Credentials conferred by for-profit colleges, 2000-2016

Image Alt Text Source: Table 318.40, Digest of Education Statistics 2017, National Center for Education Statistics. Note: Title IV institutions only.

Figure 3: Fall enrollment in non-degree-granting postsecondary institutions

Image Alt Text Source: Table 303.20, Digest of Education Statistics 2017, National Center for Education Statistics. Note: Title IV institutions only.

Figure 4: For-profit college share of enrollment, Pell Grant disbursements, and federal subsidized student loan disbursements, 2000-2016

Image Alt Text Sources: Digest of Education Statistics 2017, National Center for Education Statistics; “Trends in Student Aid 2017,” College Board.

The Rise and Fall of the For-Profit Sector

Much of the growth in the for-profit sector was driven by institutions enrolling students who were older, came from socioeconomically disadvantaged backgrounds, were veterans, or identified as a minority race or ethnicity.3 One explanation for this enrollment growth is that for-profits are relatively nimble and responsive to market needs, and therefore could attract students who were not being well served by the public and private non-profit sectors.4 The most drastic growth in for-profit college enrollment and financial aid use coincided with difficult economic conditions in the mid-2000s that weakened many families’ abilities to pay for college, led many displaced workers to seek opportunities to enhance their skills, and contributed to ongoing disinvestment in public institutions by many states.

Some also maintain that enrollment at for-profit colleges was aided by predatory marketing and recruiting practices, including falsifying financial aid forms, misrepresenting the costs and benefits of attendance, and targeting vulnerable students.5 A report released in 2012 by the Senate Committee on Health, Education, Labor and Pensions further alleged that some for-profit schools failed to properly invest in the success of students, emphasizing recruitment at the expense of student supports for program completion and post-program job attainment.6 These allegations are troubling when combined with research showing that job market returns for the average student who attends a for-profit college are often similar or worse than peers who attend lower-cost public colleges.7

The Department originally proposed the Gainful Employment rule in 2010 in response to concerns about the dramatic growth and problematic student outcomes in some quarters of the for-profit college sector. Specifically, the GE rule threatened to take away eligibility to disburse Title IV federal aid funds from for-profit and career education programs that failed to prepare their students for "gainful employment in a recognized occupation."

Prior to the introduction of the GE regulations, the most stringent student performance-based requirement related to this clause relied on student loan cohort default rates (CDR). CDR-based thresholds became relevant in the early 1990s as student loan defaults began to rise steeply, and were originally defined as the percentage of an institution’s former borrowers who default on their federal student loans within two years of entering repayment. Institutions that did not maintain CDRs below a certain threshold (less than 25% in most years) in any consecutive three-year period lost eligibility to disburse federal loans but maintained eligibility for grant programs, while institutions with CDRs that exceeded a higher threshold (40% in most years) could lose eligibility to disburse both grants and loans. In the absence of a successful appeal, institutions violating these thresholds lost access to student loans, and potentially all federal student aid, for at least the remainder of the sanction year and the following two years. Unlike GE, the CDR rules applied to institutions of all sectors, and were calculated at the level of the institution rather than the program.

The trend of two-year CDRs is displayed in Figure 5. As you can see, colleges in the for-profit sector have the highest CDRs throughout, with a steep decline in the 1990s that directly corresponds to the timing of the implementation of the CDR sanction rules. This suggests that the CDR served as an effective policy lever to help stem overall student default, a conclusion consistent with research that has been conducted using causal methods.8 There was an upward trend in the latter half of the 2000s across all sectors, though the increase is steepest in the for-profit sector. This uptick in default rates corresponded to the worsening economic conditions and disinvestment in higher education by many states described earlier, but also led policymakers to revisit measurement in the two-year CDR amid growing concerns that institutions could manipulate short-term default rates and that, in tandem, two years was too short of a time horizon to appropriately capture default—a claim supported by emerging data that many debtors were defaulting in their third year of repayment. As a result, Congress phased out the two-year CDR and replaced it with a three-year CDR starting with the 2013 cohort, meaning that the CDR would newly be calculated as the percentage of an institution’s former borrowers who default on their federal student loans within a longer time horizon of three years after entering repayment. The thresholds for sanctions were also raised: institutions with CDRs exceeding 30% for three consecutive years or 40% in any single year would lose eligibility to disburse both federal Pell Grants and federal loans. Since these changes were made, the three-year CDR has been roughly 7-8% for private non-profits, 10-11% for public institutions, and 15-16% in the for-profit sector for the 2013, 2014, and 2015 cohorts.

Figure 5: Average two-year cohort default rates by sector

Image Alt Text Source: US Department of Education.

How Federal Sanctions Affect Enrollment

To better understand how federal sanctions like those related to CDRs—and potentially the GE rule, which was not in place long enough to see its full impact—can affect student enrollment in the for-profit sector, we examined outcomes in response to the CDR regulations imposed in the late 1980s and early 1990s that led to widespread for-profit enrollment declines and closures. As context for the time period we analyze, Figure 6 displays the share of Pell Grant recipients at two- and four-year public and for-profit colleges from the mid-1970s through 2012. For numerous reasons, Pell Grant enrollment is not a perfect proxy for financially vulnerable students, though it is a reasonable systematic identifier of undergraduate college students who have relatively low levels of income and assets. We are particularly interested in these relatively low-income and -asset students since they are most likely to be sensitive to the loss of federal student aid and are also the target of policy efforts to encourage college attendance and completion.

As you can see, during the 1980s, the share of Pell Grant recipients declined at public two-year colleges (depicted by the thick blue line) as a larger share of college students shifted to the growing for-profit two-year sector (evidenced by the thick green line). While these trends are not independently probative that stricter federal oversight on student loan outcomes—in the form of the new CDR policy—caused enrollment to shift sectors, the enrollment share declined in the for-profit sector in the late 1980s and early 1990s following the implementation of the CDR rules. Subsequently, the share of Pell Grant recipients attending colleges in the public two-year sector grew once again. Relevant for current policy debates, a similar trend appears to emerge in the late 2000s, with a shift in Pell Grant enrollment from public to for-profit colleges, particularly among four-year schools (depicted as dotted blue and green lines respectively), followed by a reversal around the time when federal scrutiny of for-profit colleges ramped up.

Figure 6: Pell Grant recipient enrollment in the public and for-profit college sectors

Image Alt Text Note: Bold lines are two-year and less-than-two-year schools and dotted lines are four-year schools. Source: Pell Grant recipient data from the US Department of Education (see Cellini, Darolia, & Turner, forthcoming).

The trends displayed in Figure 6 suggest that students shifted their enrollment into the public college sector from the for-profit sector as federal sanctions on for-profit institutions came into play. In our research, we attempt to more cleanly determine the chain of cause and effect between CDR sanctions and Pell Grant enrollment. To do this, we use administrative data on Pell Grant recipients to examine what happened to the market-level enrollment in each of the public, private non-profit, and for-profit sectors when schools of different sizes were sanctioned at different times. For example, let’s say that a for-profit college was sanctioned in Fayette County, Kentucky in 1992. We compare enrollment in the county after the sanction to the enrollment before, while accounting for contemporaneous trends of similar counties. We focus our analyses on students who are likely to only seek out local college options—those who attend colleges that offer two-year and less-than-two-year credentials (though our results are robust to considering four-year colleges as well)—and approximate a student’s choice set with counties.9

To set the stage for our research design, Figure 7 depicts a simple conceptual model of why we might expect student enrollment decisions to be affected by an institution losing eligibility to disburse federal aid because of poor student loan outcomes. As shown in the first column, the precipitating event in our analysis is the school sanction, with the sanction either being imposed on a school that the student already attends (or is considering attending), or on a school that is a local competitor.

Figure 7: Conceptual model of sanctions and enrollment decisions

Image Alt Text

Unsanctioned for-profit colleges lost about 2% of their Pell enrollment when a local for-profit competitor was sanctioned

Sanctions can then affect student enrollment decisions through three non-mutually exclusive channels, as depicted in the second column. First, sanctions may impact a student’s expected net cost of attendance: the gross cost of attendance (the “sticker price”) minus financial aid. A sanctioned institution loses eligibility to disburse federal loans, and in some cases, Pell Grants and other federal aid. As a result, current and prospective students of sanctioned institutions are likely to experience an increase in their expected out-of-pocket college costs. Second, sanctions may provide new information to prospective students about a particular institution, sector (for-profit, public, or private non-profit), or course of study. By definition, sanctioned institutions have a high percentage of students who are unable to repay their student loans, which may lead prospective students to infer a lower chance of success or lower expected benefits from attendance. Third, sanctions may induce sanctioned schools to change their behaviors, such as adjusting recruitment, pricing, or aid practices. Alternatively, sanctioned institutions may be unable to support their operations if reductions in student enrollment are sufficiently large, leading some sanctioned institutions to close.

All of these changes—prices, information, and institutional behavior—have the potential to lead students to choose a different college, sector, or field of study; alternatively, students may drop out of higher education if no acceptable alternatives are available. For example, students might be able absorb higher costs if they can find additional funds by taking on relatively expensive private student loans or working more. Or students might instead decide to switch to a lower cost competitor institution or forgo postsecondary education altogether. Similarly, information about student loan repayment prospects at a college in a particular field or changes to the programs from which prospective students can choose may induce students to switch to unsanctioned competitor institutions or exit higher education altogether.

The Effect of Sanctions on Pell Grant Recipients’ Enrollment Decisions

Here’s what we found. When a for-profit college was sanctioned, annual enrollment of Pell Grant recipients at that school declined precipitously—by nearly 70%. In addition to the effect of sanctions on the sanctioned school, a key implication from our study is that reputational effects of a sanction can spill over onto institutions that students view as similar—leading students to switch to local institutions in a different sector or a different field of study. Unsanctioned for-profit colleges lost about 2% of their Pell enrollment when a local for-profit competitor was sanctioned. This might be because students downgraded their impression of other institutions in the same sector (e.g., other for-profits when a for-profit college is sanctioned). There is some question as to the extent to which students are aware of the student loan repayment or default rates at specific institutions they attend or consider attending. Information on student loan outcomes is now available on the Department’s College Scorecard website; in the past, this information was less readily available, but we found frequent mention of student loan-related sanctions when we reviewed newspaper archives. For example, after a competitor shut down in 1991, one for-profit college founder suggested that the news stories about bad outcomes in the sector had tarnished the image of the local trade school industry, which had since “taken a pretty heavy beating. Any damage to any private career school affects every other school."10

And indeed, many affected students in our study did switch their enrollment to local competitor institutions: about 60-70% of Pell Grant recipients who would have enrolled in sanctioned for-profit institutions in the absence of the regulatory sanctions enrolled instead in the public sector. Because public community colleges are typically large relative to their peers, each for-profit college sanction resulted in the Pell Grant enrollment at each unsanctioned local public community college growing by about 7% annually. Accounting for the different average size of the colleges in each sector, we found that for-profit college sanctions led to county-wide total Pell Grant enrollment declines of about 2%. We also found suggestive evidence that student loan outcomes improved in the county as students shifted from the for-profit to public sector, with the rates of borrowing and student loan default both declining.

Implications for Higher Education Policymakers

Our results suggest important implications for current debates about higher education: while regulations in many contexts do not always result in changes in behavior, our findings demonstrate that restricted financial aid availability due to sanctions on an institution does affect student enrollment decisions. These altered enrollment decisions can have downstream effects on labor market outcomes, student loan repayment, and societal welfare. Sanctions related to federal financial aid access appear to have particular bite in the for-profit sector, as we do not find similar enrollment shifts when public or non-profit institutions are sanctioned. This could be because of the relatively high prices charged at for-profit colleges, the lack of institutional aid, and the financial backgrounds of students who disproportionately attend college in the for-profit sector.11

Sanctions in the for-profit sector also had a reputational effect on unsanctioned local for-profits. The effects we found were largely exclusive to the for-profit college sector, with no evidence of significant spillover effects when a public or private non-profit college was sanctioned.

Not all students at sanctioned schools transferred to other programs: overall enrollment in higher education declined in the geographic area where that school was located. However, when for-profit colleges were penalized, most students who would have attended a for-profit institution in the absence of a sanction enrolled in public colleges. Of course, there are many important differences in the higher education market of today relative to the 1990s, including declining public support for community colleges and concerns over capacity constraints at public colleges in some states, as well as a growing number of online education options. Nonetheless, the evidence that many students shifted to the public college sector from the for-profit college sector should allay concerns that vast numbers of students who cannot access federal aid to attend for-profit colleges will not be able to find programs to fit their needs in other areas of the higher education system. Moreover, the observed enrollment shift from the for-profit sector to the relatively low-cost public sector also corresponded with improved student loan repayment outcomes. This improvement aligns with research indicating that labor market prospects and student loan repayment are often better at schools in the public and non-profit sectors than those in the for-profit sector.12

Conclusion

In repealing GE, the Department emphasized the role that transparency and public information could play in holding higher education institutions accountable.13 The idea is that such information would allow students to make informed enrollment and borrowing decisions. But while a greater amount of accessible information about programs and colleges is undoubtedly necessary, research suggests that transparency alone is unlikely to be sufficient to meaningfully improve student outcomes given the complexity of college prices and aid, especially for students that come from communities without a tradition of college-going on which they can rely.14 As federal policymakers rethink accountability structures in the wake of the repeal of GE, they must consider how best to protect students and taxpayers, and ensure that institutions are held accountable for providing value. Our research shows that sanctions work, and can be employed successfully as a lever to influence both institutional behavior and student enrollment decisions—the question for federal policymakers, then, is how and toward what goals they choose to direct such levers in future legislation.

Topics
  • Higher Education345

Endnotes

  1. The data used in the figures in this report include data on Title IV participating institutions only. Estimates indicate that there have been a substantial number of for-profit colleges that do not participate in Title IV programs (See: Cellini, Stephanie Riegg, and Claudia Goldin. “Does Federal Student Aid Raise Tuition? New Evidence on For-Profit Colleges.” American Economic Journal: Economic Policy, vol. 6, no. 4, 2014, pp. 174-206.)

  2. Authors’ calculations based on data from the US Department of Education. See: United States, US Department of Education, Office of Federal Student Aid. “Proprietary School 90/10 Revenue Percentages,” studentaid.ed.gov/sa/about/data-center/school/proprietary. Accessed 1 Oct. 2019.

  3. Cellini, Stephanie Riegg, and Rajeev Darolia. "High Costs, Low Resources, and Missing Information: Explaining Student Borrowing in the For-Profit Sector." The ANNALS of the American Academy of Political and Social Science, vol. 671, no. 1, 2017, pp. 92-112. See also: Deming, David, Claudia Goldin, and Lawrence F. Katz. “The For-Profit Postsecondary School Sector: Nimble Critters or Agile Predators?" Journal of Economic Perspectives, American Economic Association, vol. 26, no. 1, 2012, pp. 139-164.

  4. Armona, Luis, Rajashri Chakrabarti, and Michael F. Lovenheim. “How Does For-Profit College Attendance Affect Student Loans, Defaults, and Labor Market Outcomes?" National Bureau of Economic Research working paper No. 25042, Sep. 2018, www.nber.org/papers/w25042. Accessed 1 Oct. 2019. See also: Deming, David, Claudia Goldin, and Lawrence F. Katz. “The For-Profit Postsecondary School Sector: Nimble Critters or Agile Predators?" Journal of Economic Perspectives, American Economic Association, vol. 26, no. 1, 2012, pp. 139-164; Gilpin, Gregory, Joseph Saunders, and Christiana Stoddard. “Why Has For-Profit Colleges’ Share of Higher Education Expanded So Rapidly?" Economics of Education Review, vol. 45, 2015, pp. 53-63.

  5. “Attorney General Xavier Becerra Sues For-Profit Ashford University For Defrauding and Deceiving Students.” Press Release, State of California Department of Justice, Office of the Attorney General, 29 Nov. 2017, oag.ca.gov/news/press-releases/attorney-general-xavier-becerra-sues-profit-ashford-university-defrauding-and. Accessed 1 Oct. 2019. See also: Cottom, Tressie McMillan. Lower Ed: The Troubling Rise of For-Profit Colleges in the New Economy, The New Press, 2017; “For-Profit Colleges: Undercover Testing Finds Colleges Encouraged Fraud and Engaged in Deceptive and Questionable Marketing Practices." US Government Accountability Office, 4 Aug. 2010, www.gao.gov/assets/130/125201.html. Accessed 1 Oct. 2019.

  6. United States, Congress, US Senate Committee on Health, Education, Labor and Pensions. For Profit Higher Education: The Failure to Safeguard the Federal Investment and Ensure Student Success. 30 Jul. 2012, www.help.senate.gov/imo/media/for_profit_report/PartI-PartIII-SelectedAppendixes.pdf. Accessed 1 Oct. 2019.

  7. Armona, Luis, Rajashri Chakrabarti, and Michael F. Lovenheim. “How Does For-Profit College Attendance Affect Student Loans, Defaults, and Labor Market Outcomes?" National Bureau of Economic Research Working Paper No. 25042, Sep. 2018, www.nber.org/papers/w25042. Accessed 1 Oct. 2019. See also: Cellini, Stephanie Riegg, and Nicholas Turner. “Gainfully Employed? Assessing the Employment and Earnings of For-Profit College Students Using Administrative Data.” Journal of Human Resources, vol. 54, no. 2, 2019, pp. 342-370; Darolia, Rajeev, et al. “Do Employers Prefer Workers Who Attended For-Profit Colleges? Evidence from a Field Experiment." Journal of Policy Analysis and Management, vol. 34, no. 4, 2015, pp. 881-903; Deming, David J., et al. “The Value of Postsecondary Credentials in the Labor Market: An Experimental Study.” American Economic Review, vol. 106, no. 3, 2016, pp. 778-806.

  8. Darolia, Rajeev. “Integrity Versus Access? The Effect of Federal Financial Aid Availability on Postsecondary Enrollment." Journal of Public Economics, vol. 106, 2013, pp. 101-114.

  9. We use counties to proxy for local higher education markets, but results are robust to alternative definitions of local higher education markets, in which we consider an institution’s competitors to be the set of schools within a 15- or 30-mile radius. We define markets locally since the time period we study is prior to widespread broadband penetration and nearly all sanctions were imposed on institutions offering two-year (or shorter) credentials. Consequently, most prospective for-profit students’ choice sets were limited to colleges in their immediate vicinity, allowing us to capture the full set of enrollment effects in a given market.

  10. Morgan, Babette. “Vatterot College Taking Over Draughon.” St. Louis Post Dispatch, 16 May 1991, p. 6B.

  11. Cellini, Stephanie Riegg, and Rajeev Darolia. "High Costs, Low Resources, and Missing Information: Explaining Student Borrowing in the For-Profit Sector." The ANNALS of the American Academy of Political and Social Science, vol. 671, no. 1, 2017, pp. 92-112.

  12. Armona, Luis, Rajashri Chakrabarti, and Michael F. Lovenheim. “How Does For-Profit College Attendance Affect Student Loans, Defaults, and Labor Market Outcomes?" National Bureau of Economic Research Working Paper No. 25042, Sep. 2018, www.nber.org/papers/w25042. Accessed 1 Oct. 2019. See also: Cellini, Stephanie Riegg, and Nicholas Turner. “Gainfully Employed? Assessing the Employment and Earnings of For-Profit College Students Using Administrative Data.” Journal of Human Resources, vol. 54, no. 2, 2019, pp. 342-370; Darolia, Rajeev, et al. “Do Employers Prefer Workers Who Attended For-Profit Colleges? Evidence from a Field Experiment." Journal of Policy Analysis and Management, vol. 34, no. 4, 2015, pp. 881-903; Deming, David J., et al. “The Value of Postsecondary Credentials in the Labor Market: An Experimental Study.” American Economic Review, vol. 106, no. 3, 2016, pp. 778-806.

  13. United States, US Department of Education, Office of Postsecondary Education. “Program Integrity: Gainful Employment.” Code of Federal Regulations, Government Printing Office, 2019, 34 CFR 600 and 34 CFR 688.

  14. Bergman, Peter, Jeffrey T. Denning, and Dayanand Manoli. “Is Information Enough? The Effect of Information about Tax Benefits on Student Outcomes.” Journal of Policy Analysis and Management, vol. 38, no. 3, 2019, pp. 706-731. See also: Bird, Kelli, et al. “Nudging at Scale: Experimental Evidence from FAFSA Completion Campaigns.” National Bureau of Economic Research Working Paper No. 26158, Aug. 2019, www.nber.org/papers/w26158. Accessed 1 Oct. 2019; Darolia, Rajeev, and Casandra Harper. “Information Use and Attention Deferment in College Student Loan Decisions: Evidence From a Debt Letter Experiment.” Educational Evaluation and Policy Analysis, vol. 40, no. 1, 2018, pp. 129-150; Marx, Benjamin M., and Lesley J. Turner. “Student Loan Nudges: Experimental Evidence on Borrowing and Educational Attainment.” American Economic Journal: Economic Policy, vol. 11, no. 2, 2019, pp. 108-141; Oreopoulos, Philip, and Uros Petronijevic. “The Remarkable Unresponsiveness of College Students to Nudging and What We Can Learn From It.” National Bureau of Economic Research Working Paper No. 26059, Jul. 2019, www.nber.org/papers/w26059. Accessed 1 Oct. 2019.