The Importance of Stress Testing in Higher Education

This article is part of the Finance Committee Chair Toolkit, a resource that highlights a variety of perspectives, including how to assess disruptive scenarios and what types of actions to consider. It is also part of the Assessing Financial Vitality Toolkit, which includes a host of resources that enable board members to take action in ensuring institutional sustainability.

Among the most important fiduciary responsibilities of a governing board is maintaining an institution’s long-term health. Stress testing is one of the primary means of doing so.

Stress testing has been part of the ongoing operation of the human species for centuries. The objective of stress testing has been to determine if an operation can withstand shocks that are brought on by influences that may be beyond our control. Stress tests are part and parcel of many aspects of life. During a physical, a doctor stress-tests us by subjecting our body to conditions beyond what they may experience in everyday life. Stress tests have always been part of the final procedures before computer operating systems are brought online. Companies have performed stress tests on their operations for decades, and after 2008 the Dodd-Frank Wall Street Reform and Consumer Protection Act required financial institutions deemed “too big to fail” to perform stress tests on their organization under stringent oversight by Federal Reserve bankers.

It is important to distinguish between enterprise risk management (ERM) and stress testing. ERM is the process of evaluating all risks associated with an association. This includes the full gambit of the risks that might impact the institution under normal circumstances. Stress tests examine a specific set of events that might occur, but are not likely to occur, whose consequences have a significant financial impact. Stress testing is part of the ERM process.

The goals of stress testing are to determine which key aspects of an operation may be subject to failure and to attempt to establish safeguards that ensure that such stresses will not result in the failure of the entity. Unfortunately, very little stress testing is performed in the world of higher education. This creates a vulnerability for specific institutions as well as the sector as a whole. Some of these very painful outcomes were evident within the higher education sector during and after the Great Recession and some are being experienced right now.

It is no secret that higher education has entered a period of unprecedented stress. This is evident from the large number of college closures and mergers of the last several years. The marketplace and economics of the sector are changing rapidly, much of this beyond the control of any individual institution of higher learning. Most recent data show that only 34 percent of colleges met their enrollment goals in fiscal 2019. The demographics of the country have been working against the industry. Since 2011 total enrollment in postsecondary education has dropped 12 percent and the projections for the next decade do not indicate improvement, particularly in the Midwest and Northeast. Looking out a little further, there is expected to be a precipitous drop in 18-year-olds in 2026. In an article in Forbes dated December 13, 2018, entitled “Will Half of all Colleges Really Close in the Next Decade?” Michael Horn notes that “demographics are beginning to work against traditional colleges and universities. The pool of 18-year-olds is starting to decline—with precipitous declines in certain regions forecast to begin in 2026.” He also arrives at the measured answer to his question asked in the title.  “[I]t’s not hard to paint a picture of how 25 percent of existing institutions—be it 550 nonprofit and public four-year institutions or 1,100 degree-granting institutions—close, merge, or declare bankruptcy in the years ahead.”

With such dire predictions, what are the responsibilities of trustees of any institution of higher learning? Trustees are in a unique position in their organization to focus on the long term. As we all know, the administration is preoccupied with the myriad issues that confront the organization in the short term. They are worried about next year’s enrollment, next year’s fund-raising goals, next year’s budget, the pedagogy and a large number of other short-term issues pummeling them daily. Most administrators do not have the time or perspective to ponder what will happen five or more years hence. The trend toward shorter terms for presidents exacerbates this tendency.

Trustees are the only members of an institution whose most important job is to ensure the long-term viability of that institution—not one year or five years out but the truly long term. Therefore, I believe that it is the obligation of every board to oversee the performance of stress tests for their institution. While things may be going well now, that can change, and it is the duty of the board to understand where the vulnerabilities of the college may lie. The best way to do this is through stress testing.

Stress testing helps the board (and the administration, for that matter) identify which key aspects of the institution are most essential to financial viability. Colleges and universities can seem like relatively simple organizations, but they are quite complex with a number of interrelated moving parts. Things that may seem important may not be and vice versa. Stress testing can illumine a number of factors at the institution. And it is important to recognize that every college and/or university is different from all others. Stress testing can shed light on the following factors related to the organization:

• The key points of potential failure;
• The relationship between correlated components;
• The scenarios that will adversely impact the institution;
• The key components of financial viability;
• Potential hedging or mitigation opportunities;
• Training to prevent panic when things go wrong.

Through my work with nonprofits, I have observed that organizations benefit from stress tests. These nonprofits have included the Harvard Endowment, the Association of Governing Boards of Universities and Colleges (AGB), and Valparaiso University.

We started looking at stress tests for the Harvard Endowment in the early 1990s. In addition to using historical data of stock market crashes of 1929 and 1987, we also imagined such scenarios as a terrorist attack on the World Trade Center and a default by Russia on its debt. The objective was to determine how much the value of the portfolio would lose in absolute terms relative to the risk benchmarks established by the university and the impact on distributions for operations. These tests were performed every quarter (the portfolio changes over time) and helped management understand where risks in the portfolio may have been correlated, as well as the liquidity challenges the organization may face. As a result of this work, we made some changes to the portfolio and put in place a very large committed, secured line of credit.

We performed stress testing biannually for AGB. These tests were performed for the purpose of determining how the organization would be impacted by a significant challenging period of time. We used a method that I would call the “Great Recession on steroids.” This test assumed that the impact on the revenues of AGB that took place within the 2008 to 2011 time frame would occur again but would be much more severe and much longer in duration. It also evaluated the impact of very poor markets on the quasi-endowment. These tests gave the board an understanding of the challenges faced by the organization in an extremely stressful environment. As a result of the tests, the board decided to keep the quasi-endowment extremely liquid to hedge against future challenges that were unlikely to occur but would be very harmful to the organization.

Over the years, Valparaiso University developed a very robust internal financial analysis system that enables the institution to track net tuition down to the student level and provides a very flexible approach to constructing “what if” scenarios. For several years, multiple stress tests have been performed by the administration in consultation with the Committee on Finance and Administration. One very simple test that was performed (certainly not at maximum stress) was to take a one-time reduction of 100 students in the incoming freshman class. The other assumptions included zero growth in the endowment and that the discount rate would increase slightly while maintaining a gross tuition growth in line with the historical average. Another major assumption was that the university would begin to cut expenses two years out in response to the drop in revenue.

The result of this stress was that the university would run operating deficits of several millions of dollars for several years before the expense cuts and other revenues were implemented.

Since running the stress test, the board and the administration have undertaken several steps to mitigate the impact of this stress environment:

• Try to reduce expenses to levels when enrollment was lower;
• Reaffirm the focus of the capital campaign on endowment growth;
• Maintain a committed line of credit;
• Invest in student retention;
• Invest in marketing and student acquisition;
• Update the campus facility master plan to determine if any real estate is available to provide ancillary revenues;
• Create an ad hoc committee to examine other long-term approaches to improve financial sustainability.

Who Should Perform the Stress Tests?

A board of trustees is ultimately responsible for stress testing. As mentioned earlier, the board is the primary entity within a university with the obligation to work on longer-term issues, and I believe that stress testing is one of the most important responsibilities of a board of trustees. It is through stress testing that the board can much more fully understand the workings of the institution to which they have the ultimate fiduciary obligation.

To start, the board should designate a group or committee of its members to oversee the process. This can be a standing committee such as audit, finance, enterprise risk management, or an ad hoc committee, if necessary. The board should work with management to develop scenarios that might be particularly challenging to the institution. This could include such events as a significant drop in enrollment, the loss of a college or graduate school, the loss of international students, a significant change in the competitive landscape for a portion of the institution, significant change in the value of the endowment, changes in laws impacting the college, changes in such funding as Pell grants or state funding, or changes in the tax laws that may impact giving. The key is to identify those scenarios that could actually occur and would have a significant impact on the institution. The use of imagination is important in developing these scenarios. The events should be possible but not necessarily probable.

After the committee, working with management, outlines the scenarios, management should then perform the financial modeling of the scenarios and report back to the committee. The administration is best suited to perform these detailed stress tests, because they have access to the underlying data. The outcomes should then be reviewed and evaluated by the committee. If the administration does not have the resources to perform the stress testing, it should consider engaging an outside consultant.

Key Factors to Consider

While executing the testing and examining the outcomes the following concepts should be considered:

First, significant stress to an institution is typically not the result of a one-year event, so testing should examine the impact of a stressful scenario over at least three years. Stress compounds over time. As it relates to the examination of finances, the board should spend most of its time looking at multiyear projections. Annual budgets should always be accompanied by five-year projections.

Second, scenarios can be based on long-term trends that have been observed in the operation. For instance, if the number of foreign students show a slight decrease recently, the scenario should be created that imagines this trend worsens significantly.

Third, the stresses should be considered institutionwide. If one segment of the institution can be impacted by a scenario, it is very likely to impact other parts.

Fourth, stress testing is a process, not a one-time activity. As your institution performs stress tests, the administration and the board will enhance their knowledge of the process and the approach. It is an iterative activity and will improve over time.

Fifth, when the committee and administration complete the process, they should report the results to the full board.

Sixth and most important, do not let perfect be the enemy of good. It is very easy to question the impact on the institution of every assumption and to argue over details. Discussion is fine, but it is easy to become bogged down in too much debate and never complete the project. Sometimes assumptions have to be made; understand them for what they are, but do not let discussions interfere with the process of reaching definitive conclusions.

It is important for the board to understand the key components of the institution that may impact its viability. No two institutions are the same or impacted by the same factors. It is also important to be able to identify which factors correlate with one another. For instance, prior to 2008 it was believed that many of the components of revenue of a university were noncorrelated, partly because there were many years of improving trends. However, in 2008, we learned that enrollment, discount rates, state funding, endowment values, advancement, available liquidity, and ancillary services were all positively correlated. Therefore, the stress experienced by the higher education sector was more severe than anyone would have anticipated.

The financial resources map on page xx is a schematic of the operations of a relatively simple private college. While on the surface it may seem like the operation of a private college is straightforward, there are many interrelated parts that can impact the outcome of various scenarios. As you expand the operation of an institution, the complexity increases exponentially.

You may use the financial resources map to identify key areas of vulnerabilities for your institution. Each institution is different. It may be enrollment, the value of the endowment, government funding, the number of foreign students, or the expense structure. Then develop a scenario based on those areas that would have a significant impact on your institution. A good place to start may be based on history. If your institution was challenged by the Great Recession, use that information as a starting point and assume that the trends during that time were in fact worse and were longer in duration than they actually were. Alternatively, develop a scenario related to the continued demographic trends in this country, but with a regional focus.

Using your first scenario, you should then go about estimating the impact on the revenues, expenses, and balance sheet, understanding what components of the financial structure are correlated. The matrix on page xx is a heat map that can be utilized to evaluate the impact on an institution.

Those areas that are highlighted in red within the matrix have the greatest impact on the financial viability of an institution, whereas blue has the lowest impact. A ++ in the correlation box indicates the highest level of correlation, while 0 indicates the line items are not correlated. Many of the most important areas of financial viability (in red) are highly corelated with one another; while expenses are high impact, they are not correlated with revenues. In a university environment, it takes years to cut expenses as revenues drop off.

The output from the stress testing should include the key variables associated with financial sustainability, including net operating income, net cash flow, maintaining key debt covenants, the ability to maintain the liquidity needed for operations, and the impact on liquid and total reserves.

Potential Mitigation Techniques

Once the stress test is completed and the results are examined several steps can be considered to prepare for the stressful events. First, you could consider changing the endowment distribution policy to reduce spending in anticipation of stressful events in the future. In doing so the institution should examine the structure of the endowment to determine if it is possible to build or maintain unrestricted funds that could be used in a difficult period or whether a reduction in spending today could be offset by increases in the future. One interesting aspect of an endowment is that its value is often highly correlated with other revenue lines, so increased spending in a challenging environment is counterintuitive and results in selling assets at what could be cyclical lows in the market (think of 2009). One way to address this issue is to create a “rainy day fund” in which excess distributions could be held and in relatively liquid assets and invested in a way that would be less sensitive to market forces as the endowment pool.

A third approach could be to acquire a committed credit line from one or more banks. This line would be available to cover cash flow shortfalls and could be secured by unrestricted endowment assets. While there is a cost associated with having a committed line of credit, it is probably less expensive than keeping a considerable amount of cash on hand to deal with challenging times.

Another mitigation technique is to consider investing in other lines of business that may hedge the core business. For instance, we know from history that when unemployment increases, the demand for technical graduate education increases. So, if your institution is undergraduate-focused, it may make sense to create graduate and certificate programs that would likely increase enrollment when undergraduate demand is challenged.

Other mitigation techniques include examining the expense side of the income statement. We all know that Bowen’s Law, which was developed in 1980, is still alive and well in higher education. The late economist and educator Howard Bowen observed that in the pursuit of prestige, universities will spend all the money they can such that the cost of higher education will be determined by the revenue produced. Over the last 30 years revenues have increased at rates well above inflation and therefore the costs have increased at the same rate. So, we enter the current environment with a bloated cost structure. As mentioned earlier, the expenses associated with operations and physical plant are very difficult to adjust even over a two- or three-year period, so they are highly correlated when revenue is increasing and uncorrelated in the opposite environment. Therefore, one way to deal with a stressful environment is to keep a long-term cap on increases in expenses. The positive difference between revenues and expenses should be saved for the proverbial “rainy day” and not spent on new buildings or projects.

Other approaches to controlling expenses could be variable compensation schemes, such as bonuses for employees, that could be cut in difficult times or increasing the percentage of contingent faculty.

During the good times, a rubric and process should be developed for eliminating programs and shuttering facilities. The bloat of continuing with programs that are no longer in demand is a problem throughout the higher education sector.

Another approach is to examine those large expense budget items that have variability and determine whether those expenses are hedgeable or insurable. For instance, if a significant portion of the debt is floating rate, it may make sense to hedge some of the volatility of interest rates. The same could apply to utility costs.

In general, in trying to deal with the expense side of the income statement, it is very difficult to reduce expenses when dealing with a dramatic drop in revenue resulting from a series of stressful events. The key is to attempt to break Bowen’s Law by keeping expenses under control when times are good.

Stress testing is one of the most important obligations of the board of trustees of any college or university. It enables the board to take an appropriate long-term approach to the finances and operations of an institution. It provides the board with the added benefit of gaining an in-depth understanding of the organization and can help prepare the board to deal with stressful situations that may arise during a board member’s tenure. In sum, it is something every board must do whether on its own or with the help of an external party.

Verne Sedlacek is a former president and chief executive officer and board member of the institutional investment firm Commonfund and is a member of the Board of Directors of Valparaiso University and a former member of the AGB Board of Directors.