Enrollment, Tuition, and Discount Rate: The Sweet Briar Story, Part 3

This is the last in a trilogy of posts about enrollment and its role in the Sweet Briar story. The first of the series showed that there were no enrollment trends, at Sweet Briar or nationally, that pointed toward the need for Sweet Briar to close. The second of the series showed that the demographic changes in the student population that have been used to justify its closing were not a necessary result of historical trends. The revenue problems were brought about by poor administrative decision-making and an inexplicable refusal to change course. This final post will continue along similar lines but will focus on the mysterious links among enrollment, discounting, and revenue. If the last post was uncomfortable because of issues relating to race, this post is uncomfortable in that people are treated as commodities. But it’s clear that revenue matters, and in this regard all categories of students are not equal.

So, How Many Students Go to Sweet Briar?

For those who are confused about all the different enrollment numbers being thrown around, you are not alone. It is a mess brought about by the different categories of students who may or may not be included in different metrics.

For example, the Washington Post wrote: In 2010, the college had 760 students. Last fall, it reported 700.

Moody’s claims, in one of the more misleading statements among many, that from 2008-2014, enrollment dropped 15%, to 681. For them, 2010 enrollment was 743.

Inside Higher Ed reported 2010 enrollment at 605.

Then there is the Roanoke Times reporting first year enrollment for 2014 at 154, whereas Sweet Briar sometimes lists the incoming class at 169.

In another article, the Roanoke Times quote SBC lawyer Woody Fowler as saying that Sweet Briar needs 800 students but as of spring 2015 has 532.

One of the reasons we are seeing all these different numbers is that certain ones fit people’s specific agendas better than others. Moody’s, for example, seems intent on painting the worst possible picture of Sweet Briar’s situation, and they cherry pick numbers and present graphs in a way that seems designed to mislead ( an interesting coincidence is that a current SBC director is also a current and former Moody’s executive).

Another reason for all of the different numbers that have been reported is more innocent: Most people have no idea why there are multiple metrics, let alone what the differences among them are. That was the position I was in until recently. So, I have done my best to figure it out and, for the purposes of the post, to figure out what it all means for revenue.

A Case Study of 2010-2011

I have more detailed financial information for 2010-2011 than for other years. That year also happens to be key in understanding what happened at Sweet Briar. So, to understand the more general revenue issues, I will first focus on the enrollment and revenue numbers from 2010-2011.

Total Enrollment

Let’s start with the Washington Post enrollment number for fall 2010: 760. This is the highest enrollment number you will see for that year unless someone presents 12 month enrollment numbers. As far as I know, no one has done this, so we’ll stick to the fall enrollment numbers.

The 760 refers to fall total enrollment. This is the metric used by US News and World Report in their rankings. It includes: Full time degree seeking undergraduates, part time undergraduates, full time graduate students, part time graduate students, and students enrolled in Sweet Briar’s Junior Year in France or Junior Year in Spain programs. A faculty spouse who takes one class, a junior from Reed College who spends a semester in Paris in a Sweet Briar study abroad program, and a full-time on-campus psychology major all contribute equally to this metric.

As can be seen in Figure 1, of the 760 enrolled students in fall 2010, 599 were full time on-campus degree-seeking undergraduates, 24 were part time on-campus undergraduates, 9 were full time graduate students, and 4 were part time graduate students. In addition, 81 were in the Junior Year in France program, and 43 were in the Junior Year in Spain program. Add all those up and you get 760.

From FTE (Full Time Equivalent) enrollment to Headcount: the need to further separate out apples and oranges

From a revenue perspective, total enrollment is not a useful metric because the different categories of students generate vastly different amounts of money. For example, a part time student does not generate as much revenue as a full time student.

The 743 number that Moody’s uses is  FTE (Full Time Equivalent) enrollment. This number is total enrollment with a correction for the part time students. For Sweet Briar, the correction is to count each part time student as 40% of a full time student. This means that the 28 part time students reduce the enrollment number from 760 to 743 FTE, as 40% of 28 rounds to 11. FTE is the number that is multiplied by tuition/fees to determine the gross tuition revenue number reported in financial documents.

FTE might be a good number to use at other institutions, but it’s not at Sweet Briar.  Not only does it consider undergraduates and graduate students to be equivalent, but it also considers on-campus and JYF/JYS students to be equivalent. This is problematic because JYF and JYS have very different revenue/expense structures than do on-campus programs. Furthermore, the majority of JYF/JYS students are students from other institutions taking advantage of Sweet Briar’s program. Adding on-campus and JYF/JYS students together combines apples and oranges.

So, let’s break our FTE number into its component categories and look at the differences in revenue obtained from the different types of students. If we separate out the full time and part time undergraduate on-campus students from the FTE number of 743, we get 609. (the 24 part time undergraduates correspond to 10 full time undergraduates). Tuition for each of those 609 FTE students was $29,245. In addition, each of the 599 full time students paid another $475 in fees.

For the 13 graduate students, the situation is different. All tuition is generated on a per credit hour basis, and it differs for the two graduate programs Sweet Briar offers, one leading to a Masters of Education (MEd) and one to a Masters of Teaching (MAT). For the MAT program, the 2010 tuition was $475/credit hour. For the Med, it was $310/hour. In fall 2010, there were 5 full time MAT students, 1 full time MEd student, and 3 part time MEd students (those students were apparently counted as full time students in the IPEDS database, but Sweet Briar internal documents have them listed as part time students). Total net revenue was a year equivalent of $73,200, or, using the part time = 0.4*full time conversion, approximately $10,000 per FTE student. In addition, there were 4 part time non-degree seeking students who brought in a year equivalent total of $7560, less than $2,000/student. In total, then, the 13 graduate students brought in a total of a year equivalent $80,760. In 2010, it appears that the average part time graduate student was only worth about 0.2 full time students, leading to a graduate student FTE of 7.4. This corresponds to about $10,913 per student.

So, the best guess is that each FTE graduate student is worth a little more than a third of a FTE undergraduate, that each part time graduate student is worth a fifth of a full time graduate student, and that each part time undergraduate is worth two fifths of a full time undergraduate. The small sample size means that these are rough estimates, but they are close enough for current purposes.

This has probably already gotten confusing, and I’ll summarize soon, but first, there is one more large group to address – the Junior Year in France and Junior Year in Spain students. In fall 2010, there were 81 JYF and 43 JYS students. Based on Sweet Briar budget documents, if those numbers were the same in the spring, revenue from the programs would have been $3,145,430 and $1,393,050, respectively. This means that each additional JYF student brought in $38,832/year-in-program, and each additional JYS student brought in $32,397/year-in-program.

Here’s a summary of where we are. In 2010, in terms of gross revenue from tuition and fees, each of the following types of students contributed roughly the following amount:

• Full time on-campus undergraduate: $29,720
• Part time on-campus undergraduate: $11,698
• Full time graduate student: $10,913
• Part time graduate student: $2183
• JYF/JYS student: $35,500

One problem with knowing what to make of these numbers is that room and board is included in the JYF/JYS revenue (there is no separation of tuition from room & board), and expenses for junior year programs are different and separate from other Sweet Briar expenses. For JYF/JYS, it is a bit difficult to know exactly how to determine the proper amount of net revenue generated per additional student, as it varies year to year and does not fully scale with changes in enrollment, but a reasonable estimate is that each additional student in one of these programs nets Sweet Briar approximately $3500/year ($1750 for each semester they are abroad).

Other corrections are needed, as well.  Full time on-campus students pay room and board, but part time students probably do not. How much net revenue (revenue after expenses) is obtained from room and board? This, too, is difficult to ascertain with precision, but one can get an idea by looking at the revenue and expenses involved in auxiliary services. Besides room and board, auxiliary services also includes the conference center, faculty and staff housing, summer programs, and a number of other components to running a college.

Figure 2 shows a breakdown of auxiliary services expenses for 2009-2010 and what was budgeted for 2010-2011 (from an internal Sweet Briar document). Looking at that, it seems reasonable to assume net revenue from room and board to be around 45% of the actual cost. That is about $5000 for 2010-2011. So I’ll assume that Sweet Briar received around $5000 in additional profit from each full time on-campus student that it likely did not get from part time students and that it definitely did not get from JYF/JYS students. Although not all full time students live on campus, most do, so I won’t try and adjust further for that.

So, can we just add this $5000 to the tuition for each full time student? Yes, but first we need to know more about tuition because most students don’t pay the full tuition sticker price. For example, in 2010-2011, the average full time undergraduate paid tuition of around $15,500. It’s hard to know what part time undergraduates paid. It’s probably not the full 40% of $29,245 ($11,688) each, but is it as low as 40% of the $15,500 ($6200)? Because part time students include faculty, staff, and relatives of faculty/staff who get free tuition, we’ll go with the low-end number of $6200.

Although we do not know the exact numbers for each category of student, we’re getting closer to understanding things from a revenue perspective. Net revenue per student, which includes estimated net (not gross) revenue from room and board, when applicable, was close to the following in fall 2010:

• Full time on-campus undergraduate: $20,500
• Part time on-campus undergraduate: $6200
• Full time graduate student: $15,913 (assuming they live on campus)
• Part time graduate student: $2183
• JYS/JYF student: $3500 (year long equivalent)

(Note that that Net Tuition Revenue and Fees number reported in IPEDS and elsewhere is a bit different than what we are computing here. Those numbers are based on what the total cost to an average student receiving financial aid is and so they include the full room and board amount. Here, we are interested in the revenue to the college, so we are attempting to remove the cost of rooming and boarding from our estimates. We are also computing a number for all students, regardless of whether or not they receive some form of institutional financial aid.)

The greatest uncertainty is in revenue from graduate students and part time students, but they contribute a relatively small amount to the total revenue, so if I’m off by a bit, it won’t change the overall story. Perhaps here it’s best to step back from the details a little. In general, between graduate students, part time students, and JYF/JYS, Sweet Briar seems to net about $600,000 – $1,000,000 in revenue. The rest is from the full time degree seeking undergraduates, and it’s clear that a disproportionate amount of revenue comes from this group. For this reason, the most important metric for understanding Sweet Briar’s finances is the number of Full time degree seeking undergraduates. This measure, or something very close to it, is referred to as headcount. It is the measure that corresponds to the 605 value reported in Inside Higher Ed (small variations in this number are to be expected, as it varies depending on when, exactly, enrollment is measured). It is also the measure that VP of Finance Scott Shank used in revenue and enrollment projections that convinced the board to close. I’m not surprised he used the most appropriate metric. Shank knows numbers and how to use them effectively.

I hope it’s now clear why total enrollment and FTE enrollment are not useful metrics for understanding the revenue situation at Sweet Briar. Each additional full time undergraduate generates around 6 times the revenue per year of an additional JYF/JYS student. And each full time undergraduate may stay for four years, whereas most JYF/JYS students are part of Sweet Briar for only a semester.

Moving forward from 2010-2011

Discount rate and tuition revenue

Ok, now we are ready to tackle the issue of discount rate and what it means. We won’t have to worry about JYF/JYS and graduate students because we will be tallying the discount rate for incoming undergraduates, and incoming freshman do not participate in junior year abroad programs.

Discount rate is most commonly computed as total institutional aid divided by gross tuition/fee revenue. Institutional aid is any sort of scholarship an institution grants, whether merit-based, need-based, or other, but does not include federal and state grants. In effect, institutional aid is just a way to charge different people different amounts of tuition. Almost everyone gets some form of institutional aid and so a discount.

To figure out gross tuition revenue for an incoming class, all one has to do is multiply the number of students in the incoming class (number of freshman) by the tuition/fee sticker price. In the IPEDS database, the incoming class size relevant to discount rate is known as the Fall Cohort. For 2010-2011, this number was 180, and tuition/fees was $29,720, so total gross tuition revenue was $5,349,600. IPEDS provides numbers for total institutional aid (for incoming class) but only from 2008 on; however, we can figure out total institutional aid dating back to 2001 by multiplying the average institutional aid by the number receiving aid. Keep in mind that average institutional aid is only calculated for those students who receive aid and not for all students. This is why we can’t simply divide average institutional aid by tuition to determine the discount rate.

Figure 4 shows the discount rate since 2001 at Sweet Briar and two comparison groups of colleges. The Annapolis Group is a group of roughly 125 liberal arts colleges of which Sweet Briar is a part. It is generally considered Sweet Briar’s peer group. The other comparison is with the 42 women’s colleges for which I could find data on the IPEDS government site.

Note that although discount rate is increasing at Annapolis Group schools and at women’s colleges, the trend is smooth and gradual. That is not the case at Sweet Briar, where there was a sudden jump between 2009-2010 and 2010-2011. This corresponds to implementation of President Parker’s  recruiting/financial aid strategy.

At this point, we can figure out the average tuition revenue per incoming student by multiplying the discount rate by tuition. Figure 5 shows the inflation adjusted average tuition revenue per student. Keep in mind that room and  board are not included in this graph!

Two things jump out. First, net tuition revenue per student has been mostly flat across the sector for a number of years, even though the discount rate has been increasing. This is the result of the game colleges play in which they simultaneously raise both tuition (see Figure 6) and discount rate. They are saying, in effect, “Look, I’m a really expensive college, but I’m going to give you a great deal!” I think this is a bad approach because people may be put off by a high sticker price before they know about the amount of discount, but it’s not an unusual strategy.

I hope, though, it’s clear that students are not paying less and less to attend college at Sweet Briar’s peer institutions. But they are at Sweet Briar. Note the sharp decrease in revenue per student that coincided with Sweet Briar’s increase in discount rate (Figures 4 and 5). An inflation adjusted drop of $4750 in a single year is enormous, and for 2013-2014, there was another drop of about $2000.

When headcount is 600, a drop of $4750 per student corresponds to a drop of $2.85 million in revenue. A drop of $6600 per student corresponds to a drop of just under $4 million in revenue. The reason a drop that large in net tuition revenue wasn’t immediately observed at Sweet Briar is that the discount rate is applied to an incoming class and typically follows that class through. This means that Jo Ellen’s recruiting decision would take four years to have its full effect. But the business model to which she switched Sweet Briar would, without change, lead to a sustained $3 million/year drop in net revenue. It did change, though. In 2013-2014, the discount rate increased again, giving Sweet Briar a business model that involved a sustained $4 million/year drop in net tuition revenue as compared to the model in effect in 2009-2010.

Figure 7 shows what annual tuition revenue from full time undergraduates would have been if all students paid the same tuition as the average student in the entering class. It also projects into the future using VP of Finance Shank’s projected enrollment numbers and using the discount rate established in 2013-2014, the last point for which I have a specific value.

The administration will argue that the high discount rate was necessary to attract students (students just wouldn’t come otherwise), but what these graphs show is that this is not what happened. The large and abrupt shift between 2009-2010 and 2010-2011  is strong evidence that the revenue decrease in recent years is the result of decisions and not trends. Students don’t just stop wanting to come to Sweet Briar overnight, and it’s clear by looking at the comparison colleges that nothing special happened between 2009-2010 and 2010-2011 that would profoundly change the landscape. Rather, President Parker (presumably  in consultation  with the board) made a deliberate choice to change how Sweet Briar recruited students and how it gave financial aid, a choice that ending up costing a tremendous amount of money. I should add that most schools give most students some amount of financial aid, but Sweet Briar went all-in starting in 2010-2011. Figure 8 shows the percentage of students receiving financial aid. Since 2010-2011, only a few (if any) incoming students have paid full tuition. In 2010-2011, there were 3 who did. The previous year there were 21. Once again, this is not because of trend or inevitability but because of decision.

The projections that closed the college and what they say about leadership

The drop in revenue per student is why Sweet Briar has kept enrollment numbers up but lost large amounts of revenue. But Scott Shank’s bleak projections into next year and beyond were not just about the discount rate. He also projected headcount at 517 and 504 for 2015-2016 and 2016-2017, leading to stunningly low expected tuition revenue: $6.8 and $6.7 million. For 2015-2016, that meant that each student would be expected to pay approximately $13,150 in tuition/fees. Sticker price, which goes up every year, was around $34,900 for 2014-2015.

The revenue per student number is problematic in its own right, but the decrease in enrollment makes the problem worse. The estimated 504 students enrolled for 2016-2017 would be an exactly 100 student drop from when President Parker started in 2009-2010. When estimated net revenue from room and board is included, a drop of 100 students is a loss of about $2 million dollars.

A headcount of 504, if it were realized, would be the lowest on record dating back to 1980, when enrollment numbers are first available to me. As national trends neither move that fast (fall 2013 undergraduate headcount was 594) nor are moving in that direction, it is clear that the projections of plummeting enrollment at Sweet Briar are based on factors specific to Sweet Briar. In Shank’s model, they are based on having the same incoming class sizes as in 2014-2015 – 154 students – and Sweet Briar’s typical levels of retention. Perhaps this pessimism in the projections is warranted. It is very likely that the small class size in 2014 is due to having the Chief of Staff act as a Dean of Enrollment. Shank’s projections suggest that this lack of a qualified Dean of Enrollment was expected to continue into the future, leading to historically low enrollment numbers.

I want to emphasize this last point, because it is important. There must be a reason to expect a loss of almost 100 students, to historically low numbers, in the span of just 3 years. There must be a reason to think that the small fall 2014 incoming class was not an anomaly but signaled a new reality. If it were an anomaly, then President Jones’  statement about needing an additional $250 million in the endowment to achieve sustainable wouldn’t make sense by even his own peculiar logic. Leadership thought something changed overnight, so what was this change that led to a new reality? The only answer that makes any sense is that the new reality involved a continued refusal to hire a Dean of Enrollment. In other words, what the projections strongly suggest is that leadership knew of the problem of not having a qualified Dean of Enrollment and yet decided to model the future as though that problem would persist. Perhaps nothing shows how leadership failed to devote sufficient resources to the mission of the college than this. Not only did failing to fill a key position lead to a problem, but this lack was expected to continue into the indefinite future – as though that is an acceptable way to run a college. Well, it is – but only if the goal is to stop operations.

The Updated Sweet Briar Story

I think the Sweet Briar story is becoming increasingly clear. President Parker took over in 2009 and changed the business model in her first year to one that would lead to a sustained $3 million/year loss in tuition revenue without increases in enrollment. Perhaps it can be argued that this was a reasonable gamble, but what was not reasonable is how long the strategy persisted when it should have been clear in no more than 2-3 years that it was failing. Yet, that business model is the one used for projections going through 2016-2017. The failure to change course is more than just a mistake or lack of oversight. A nonprofit board entrusted with a college cannot act so irresponsibly.

And there was further irresponsibility. Besides implementing a failed recruiting/financial aid model, the only other substantial changes since 2009-2010 involved failures to fill the most important positions for enrollment and fundraising. The lack of a Dean of Enrollment seems to cost around $2 million/year. The lack of a Vice President of Development gives credibility to projections of historically low alumnae giving (even though alumnae giving for 2013-2014 is listed by Sweet Briar as totaling $11 million!). There is also what I estimate to be an additional $1 million/year in expenses related to interestingly timed and surprising raises in 2014 and 2015 (a 3% raise went into effect January 1, 2015 – one that won’t cost them much but helped make the future projections even bleaker).

When does mismanagement become fiduciary irresponsibility? We have to be getting close. Because what we have is a case where an institution had a viable business model, where that model was replaced by a model that was not viable, and where the caretakers of the institution closed the college in large part because they refused to change the failing model.

Loose ends

I hope that one more piece of the story is now clearer than it was and that issues relating to tuition revenue, discounting, and enrollment are now better understood. There are just two more loose ends.

1. Why is the fall 2014 incoming class sometimes described as having 154 students and sometimes 169? In the latter case, transfer students are included. For the record, it appears that, on average, a transfer student is worth a tiny bit more than half of what an incoming freshman is worth in revenue because they do not attend Sweet Briar for the full four years. Transfer students are included in the headcount number, so they’re not being ignored, but they are not counted in the fall cohort number that determines discount rate.
2. Why does SBC attorney Woody Fowler say that Sweet Briar currently has 532 students but needs 800? There are two issues here. The first has to do with the 532 number. Currently, it is the spring semester. Many first year students who transfer out do so before the start of the spring semester, so this number likely is a reflection of the 75% average retention interval for first year students. The loss of 37 students by this point seems a bit high but not too far from what I would expect based on a starting total class of 569. Attrition is just part of things. But this number is misleading because it’s compared to an 800 number that, if it had any meaning, would apply to fall numbers. The 800 number does not have any meaning, though. Even Scott Shank’s slides meant to paint the situation as being as dire as possible had the target enrollment this year at under 750, peaking at 755 over the next two years. And that number was never realistic. Did anyone really think there was a chance of increasing enrollment to unprecedented levels when the most substantial changes that took place, outside of an attempt to increase diversity, involved losing key personnel and not replacing them?
   And to emphasize just how absurd it is to reference needing 750 students, let alone 800, the final graph is of maximum occupancy at Sweet Briar. That’s right, if Sweet Briar got to even 650 students, there would be problems, and 700 would clearly take another building.