The Wrong Thinking about Measuring Costs & Efficiency in Higher Education (& how to fix it!)

There is a movement afoot to reduce the measurement of the value of public institutions of higher education to a simple ratio of the revenue brought in by full time faculty members divided by the salaries and benefits of those faculty members. That is, does each faculty member “pay” for him or herself, on an annual cash flow basis?[1]

Even some of the finest major public colleges and universities have recently succumbed to reporting such information, arguably, in an effort to appease politically motivated critics.[2] This seemingly simple ratio of the “net cost” of faculty salaries and benefits is presumed representative of the relative efficiency of higher education institutions and/or entire public systems of higher education.

This is a dreadfully oversimplified if not simply wrongheaded approach to measuring the cost of providing public higher education.  It is also a simply wrong approach to characterizing the efficiency of production of higher education institutions or higher education systems, largely because the approach ignores entirely the question of what higher education institutions produce. More importantly, measuring institutional performance and efficiency in this way does little or nothing to inform policymakers or institutional leaders on how to get more bang for the buck from higher education. That is, how to generate greater economic benefit to the state or society as a whole, by achieving more efficient production of an educated citizenry.

Arguably, the greatest economic (setting aside cultural and social) value-added of public higher education systems is achieved when those systems can efficiently transform high school graduates into college graduates, with all of the economic and societal benefits bestowed on them (at least in relative terms). This is especially true for high school graduates from low-income backgrounds, including first generation college students. Accepting an economic emphasis, public higher education institutions can and should substantially improve the economic outlook and lifelong earnings of students who otherwise have the least likelihood of college degree completion. Thus, public higher education’s role in providing value added to the economy and to society as a whole.

As such, what we must begin to better understand is how colleges and universities can improve the efficiency with which they produce undergraduate (and graduate) degrees across a variety of fields, and for students of varied backgrounds. Further, we must establish metrics of cost and efficiency that promote the right incentives for faculty and institutions of higher education to improve degree production, especially for those students previously least likely to complete their undergraduate education in a timely and efficient manner. The current policy rhetoric and proposed metrics do little or nothing to advance these policy objectives.

Flawed Reasoning and Bad Incentives of the Net-Value Approach

Under the politically popular model of faculty “net value,” the basic underlying assumption is that higher education faculty are worth as much as the sum of a) the grant funding they bring to the institution and b) the number of student credit hours they produce, thus generating tuition revenue. It is then assumed that if the state subsidized portion of the faculty member’s salary is greater than the sum of the other two values, that faculty member is inefficient (or not worth it).  Therefore, the incentives for any faculty member formally evaluated or even informally characterized by this model are to either, track down enough external grant and contract funding to pay in full, his or her own salary and/or to teach enough large sections of large classes and recruit enough students into his or her classes to cover salary and benefits. The same incentives similarly apply to all faculty. But both are counterproductive incentives.

If the mission of public higher education is to produce an educated citizenry that contributes to the economy and society as a whole, as well as being a direct engine of economic development through research and scholarly productivity, then having all faculty focus their efforts on chasing external funding to cover their costs and reduce or eliminate teaching from their responsibilities is counterproductive.  Second, production of credit hours and generating tuition may also operate at odds with helping college students progress most efficiently toward degree completion. Maximizing course enrollments generates tuition and credit hours, but may actually reduce time-to-completion as more students get lost in the shuffle. It also reduces the incentive to provide lower enrollment higher level courses that may improve completion rates.

The net-value metric is at best neutral to whether institutions try to move students forward toward completion, or allow them to flounder, repeat numerous (large enrollment) courses and never quite reach the end goal. That just doesn’t make sense, on many levels.

Finally, using this net-value metric forces the same incentive structure onto all faculty members uniformly, encouraging them to act as autonomous agents choosing either one or other approach to covering their margin.

Understanding the Role of Student Behaviors

How might we better think about productivity and efficiency in higher education? Again, consider that a primary goal is the efficient production of degreed or credentialed graduates. That is, taking high school completers and moving them efficiently through their coursework to degree completion, at which point they are likely to, at the very least, be a higher wage earner than they otherwise might have been, and in an even better light might be more likely to contribute more significantly to the economy and society as a whole.

Higher education institutions consist of a maze of pathways often navigated naively (or at least irregularly) by college students trying to find their way toward that light at the end of the tunnel. Evaluating the relative efficiency of higher education institutions requires that we better understand these student behaviors – student course taking patterns – and figure out a) which behaviors seem to be more (and less) associated with successful degree completion and b) whether institutional constraints or supports make any difference. It is naïve, if not completely ignorant to try to evaluate the productivity or efficiency of higher education systems and their economic contributions (or financial drain) without considering these student behaviors and how to influence them.

On the one hand, understanding student pathways helps us understand who is more likely to complete their degree in a timely manner. Further, for those critics of higher education who believe that too many students are pursuing (or at least completing) “useless” degrees in “unproductive” fields, it is important to understand how and why students migrate across degree programs through course selection behavior.

For example, let’s say that we believe society needs more electrical engineers than economists, a reasonable assertion indeed! (note the old adage that majoring in EE [electrical engineering] refers to “eventual economics”). Evaluation of course taking behaviors may reveal that many EE majors become economics majors (without really wanting to) after performing poorly in specific lower level engineering courses, for a variety of reasons. It may be that these students would still have been great engineers and would have flourished in their higher level courses. But perhaps course delivery approaches (large lectures) lack of supports or other institutional barriers are partly at fault.  Identifying these barriers and shifting institutional policies may lead to an increased production of electrical engineering completers (and most importantly a decrease in future economists).

Linking Student Behaviors to their Cost & Efficiency Implications

Building on understanding student pathways, we should shift our focus toward the way groups of faculty members and the sequences of courses (and degree programs) they provide lead to differences in the likelihood of degree completion, differences in time to completion and differences in the total costs of degree completion.  This is another area where higher education cost research has gone awry in the past. One cannot calculate the differences in costs of producing an economics versus an engineering major by simply looking at the costs of operating those departments. Departments are top down organizational units of universities. But students pursuing a degree in any one field take courses across many units. Instead, we can estimate the cost per credit hour for any one student taking any course in the university, and can then estimate the cumulative costs of common student pathways, and identify the higher and lower average and total cost pathways toward achieving any one degree.

Taking this approach, we might find, for example, that offering smaller class sizes (thus higher unit cost) in specific lower tier courses decreases the likelihood of repeating those courses and/or increases likelihood of successful completion of subsequent courses, leading to an overall more efficient pathway to degree completion.  But under the current model of evaluating the net cash value of faculty, the incentive works in the opposite direction by encouraging filling seats over completing degrees and programs.

We might find that offering additional supports for students from disadvantaged backgrounds (who attended high schools with weaker math and physical science programs) taking their lower level courses in engineering calculus leads to greater likelihood of timely degree completion in electrical engineering. Further, that doing so significantly decreases average cost to degree completion by decreasing course repeats.  Again, the current net-value approach creates the opposite incentive, favoring course repeats to beef up credit hour production in high enrollment lower level classes.

In reality, the unit costs of any single course, or net value of the faculty member delivering that course, matter far less than how that course more broadly influences the cost of degree completion overall.

Institutional and Public Policy Implications

For progress to be made in the current policy conversations around higher education costs and efficiency, we must improve our metrics and must link new metrics to a much deeper understanding of just how higher education systems work, the role of individual student behaviors and the complexity of the delivery systems and institutional structures designed to serve those students.

We must also be cognizant of the fact that higher education systems are not uniformly, as often characterized in policy rhetoric, stagnant structures of ancient origin, assuming a single woefully inefficient, exorbitantly costly and arcane governance and program delivery structure. Arguably, many elite institutions which best fit this caricature (elite private liberal arts colleges), while sustaining themselves with very high tuition, also achieve very high degree completion rates, albeit for the most advantaged high school graduates.

By contrast, in recent decades we have seen a dramatic proliferation of alternative delivery mechanisms, including rapid expansion of online and for profit higher education institutions. Further, many of these alternative delivery institutions have begun to disproportionately serve high school graduates with the least likelihood of timely (6 year or less) degree completion and have done so at substantial public expense through access to federal student loans. If evaluated on a net-value of faculty basis, these institutions likely look quite good. They must in order to achieve their desired financial bottom line. Yet, their financial bottom line (and in some cases stock value) comes at the taxpayer expense of high rates of loan default and societal and economic expenses of dismal rates of completion of meaningful degrees or credentials.

Getting higher education cost and efficiency measures right is critically important for informing the policy debate and for informing institutional practices. Getting these measures right means the difference between incentivizing non-productive course credit and financial debt accumulation versus incentivizing timely degree completion. When one group of students completes their degrees in a timely fashion, institutions have more resources available for the next wave. Finally, getting these measures right means the difference between a) having each and every faculty member in public institutions of higher education operate autonomously and inefficiently out of self-interest, often to the disadvantage of their students, or b) having faculty working collectively with colleagues and their institutions to improve degree production for the benefit of students, and the broader economy.