The Objectives: What Gets Measured, Gets Done

If it is true that what gets measured gets done, then we need to figure out what we want to get done. First and foremost, a federal accountability system must assess risk and return on investment on its two primary and largest student aid programs, student loans and Pell Grants. While it could be argued that loan metrics could be used as proxies for both programs, as Pell Grant recipients represent nearly 90 percent of undergraduate loan defaulters,¹ this would ignore the fact that 44 percent of Pell Grant recipients do not borrow federal student loans,² while three out 10 non-borrowers received a Pell Grant.³ Pell is the bedrock of the nation’s student aid system and reflects the federal government’s historic role of promoting access for lower-income students in higher education. Unlike unsubsidized loans, Pell is very well targeted, as 68 percent of recipients have incomes under $30,000.⁴ A balanced approach would hold schools accountable for both federal student loans and the Pell Grant.

Second, completion is a key objective and priority, both from the student and government perspectives. The relationship between completion and student loan defaults and repayment has been well established. Non-completers are three times as likely to default on their loans compared to graduates⁵ and completers are at least 20 percentage points more likely than non-completers to pay down loan principal.⁶ Students also enroll and pay tuition and other college costs with the expectation that they will earn credentials; federal requirements provide federal aid eligibility only to students who are degree- or certificate-seeking. This fundamental public policy and individual objective is also reflected in governors’ commitments to the completion agenda and is featured prominently in state OBF systems, almost all of which include completion metrics. It is worth noting, however, an unintended consequence of this metric is some institutions, especially at the two-year level, may shift their offerings to certificates and shorter credentials that have higher completion rates but less value in the labor market. Accreditation and state oversight are critical in preventing such gaming. These dangers underscore the need to avoid a single metric federal accountability system.

Third, equity is another key federal objective and the primary impetus behind the HEA, which reflected the ambitious vision of college for all.⁷ However, while expanding access to higher education has been the role of the federal government, the need to move beyond access to success is imperative given the detrimental outcomes of dropping out for student loan borrowers.⁸ Low-income students are the priority population in the HEA and knowing whether federal programs are actually increasing the degree attainment rates for that population is essential. As a result, measuring this population’s access and success rates is critical for advancing equity. It is not surprising that states have also placed equity at the core of their OBF systems, along with completion.

Fourth, the growing interest in labor market outcomes, including employment, earnings, and job placement rates, cannot be ignored and research shows that even when imperfect such data can provide valuable information distinct from students’ academic outcomes, especially for two-year and non-degree institutions.⁹ At the same time, the research shows that the simplest labor market metrics are unreliable, especially when viewed in the near-term immediately following separation and cautions against reliance on any single metric. The low rate of usage in state OBF systems also underscores the limited feasibility of a large-scale federal consequential accountability metric, at least immediately. Since getting a job is a primary objective of prospective college students,¹⁰ labor market outcomes must certainly be provided for transparency. But given data quality and methodological concerns, it is more appropriate to initially rely on other metrics that serve as proxies for labor market success for consequential accountability purposes, such as the loan repayment rate, and include labor market data solely in report-card accountability while continuing to explore a reliable metric for consequential accountability.

Finally, report-card accountability must include fully disaggregated data, in order to allow for the analysis and identification of problems that are masked by aggregate performance metrics. This is critical for equity and has been tried and tested in K–12 education with great success, as it is the one area of clear consensus that emerged from the elementary and secondary accountability movement.

In summary, a new federal accountability system should protect federal investments and promote completion and equity through institutional-level consequential accountability, and strengthen institution- and program-level report-card accountability through the addition of labor market outcomes. At the same time, existing guardrails for higher-risk institutions and programs should be maintained, with flexible opportunities for high performers.