How to find out how much a college spends per student?

College Search & Selection
21

<p>I ran a test report -- exporting total operating expenditures and FTE enrollment for JY2003 to Excel. Here are the calculated per student expenditures for a few selected LACs. These are consistent with similar numbers I had generated from the FY2004 financial reports for some of these schools:</p>

<p>WELLESLEY COLLEGE $67,020
SWARTHMORE COLLEGE $65,546
WILLIAMS COLLEGE $60,957
AMHERST COLLEGE $55,421
BOWDOIN COLLEGE $54,579
SMITH COLLEGE $53,440
POMONA COLLEGE $53,286
HAVERFORD COLLEGE $50,824
WESLEYAN UNIVERSITY $48,912
BRYN MAWR COLLEGE $46,552
DAVIDSON COLLEGE $44,363
WASHINGTON AND LEE $39,934
BARNARD COLLEGE $39,110</p>

<p>Now, obviously you have to use some common sense in looking at this data. For example, if I were investigating Barnard, I would consider that the spending may not fully reflect the resources available as part of Columbia University. The same thing may apply to Pomona where support functions like security, health center, libraries, and dining are shared with the other Claremont Colleges.</p>

22

<p>
[quote]
to get the total operating expense for each school, then use Excel to calculate per student spending (you can export the proper table from NCES to a comma-delineated file).

[/quote]
</p>

<p>The same complaints you level against the NCES data, however, can surely be leveled against the crude measure of "Operating Expenses Divided by FTE or FYES." </p>

<p>Under this measure, all spending is equated with quality, without any judge of its appropriateness or prudency. </p>

<p>The "per student" measure is a difficult and flawed way to look at schools (however one does it), which is why I assume the OP suggested (in vain) that none of us tackle that angle of it. I'm not convinced that we should privilege one method so much over another, when it's pretty arguable that both may mislead.</p>

23

<p>It is far more useful to ask what a college offers that a given student cares about, than to try to figure out how much it spends on such things. The variations in administrative structure and hence accounting make dollar comparisons near meaningless. Even if you could make a comparison, there is no reason to care. </p>

<p>If you care about philosophy, don't ask how much money they spend heating the philosophy building in the winter. Ask who teaches the courses, how well, how many professors, how many students, and other things that actually matter.</p>

24

<p>
[quote]
The same complaints you level against the NCES data, however, can surely be leveled against the crude measure of "Operating Expenses Divided by FTE or FYES."

[/quote]
</p>

<p>I don't have any problem with the NCES data. That's where the list I posted above came from. What I have a problem with is selecting one or two categories from the NCES data when there are significant variations in what expenses each school lumps into those particular categories.</p>

<p>As I have suggested, the BEST approach is to go directly to the school's own management discussion of their financials, their accreditation self-studies, and their strategic planning documentation. This will provide valuable information on what the school considers to be its strengths and weaknesses. Are they in financial equilibrium? Or are they planning to reduce the size of the faculty? Cut back on financial aid discounting? Are their salaries competitive to attract top faculty? What are their infrastructure short-comings?</p>

<p>The goal of researching any college is to determine:</p>

<p>a) what do they really charge?
b) how much do they spend?
c) what are the spending priorities?
d) do those priorities match the student's interests</p>

<p>For example, I might find that one college spends three times as much on varsity athletics as another college. Is that good or bad? Depends on the student's interests.</p>

<p>My only iron-clad rule is that, if everything else is a tie, go to the school with the larger per student endowment. Follow the money.</p>

<p>The other aside is that looking at financials for universities is 100% useless as far as evaluating undergrad education. Not because spending isn't important, but because it's impossible to isolate undergrad spending.</p>

25

<p>
[quote]
I don't have any problem with the NCES data.

[/quote]
</p>

<p>Well, that's reassuring to hear. However, please understand that you give people a different impression when, in earlier posts, you describe it as "not correct," "too limited," and "problematic."</p>

26

<p>Interesteddad,
Thanks for working the numbers, could I persuade you to extend your analysis to some of the other LACs in the USN&WR rankings? (please oh please everybody lets not digress on why USN&WR rankings are nogoodanyway).
Carleton, Middlebury, Claremont-McKenna, Davidson, Vassar, Grinnell, Hamilton, Harvey Mudd, Reed, Colby, Bates, Macalester???</p>

27

<p>Since this method is apparently appealing to justaparent, let me add some other caveats to this:</p>

<p>
[quote]
Now, obviously you have to use some common sense in looking at this data. For example, if I were investigating Barnard...

[/quote]
</p>

<p>Other common sense things: </p>

<p>Keep in mind that if the operating budget reflects deficit spending, then that rate of spending per student may not be sustainable without some kind of change in the future, either an increase in debt service or a cutting back in expenditures. </p>

<p>Also keep in mind that sometimes half of the operating budget at these kinds of institutions go to salary and benefits. Thus, variations in cost of living and other factors can mean that an institution spends a lot less on these items while still maintaining the same number (and quality) of faculty/staff/admin per student. That can have a real impact on the operating budget, and it would be insupportable to argue that spending more (or less) for these reasons reflects a difference in quality of student experience. Similarly, if a very well-managed and prudent institution found an excellent health plan for its faculty, expenses could decline. Again, it would be unfair and incorrect to say that school was "lower quality" than the institution who was still over the barrel and paying much more for benefits. </p>

<p>I also think debt service is worth looking at, particularly in terms of what it is being spent for. If it is spent on things that students are still benefitting from, then it's a non-issue. If it's spent on things that are no longer an identifiable benefit, well, that's falsely inflating the "spending per student."</p>

<p>Other expenses are also worth examining. If you're burning natural gas instead of heating oil in the buildings, and maintaining the exact same temps for the students, the cost difference between one and the other this winter means it's more expensive per student on one campus--but it feels funny to imply you're "spending more" on a better undergraduate education.</p>

<p>Maybe it's the higher ed budget nerd in me, but honestly? I think a more realistic way to frame it is to call this EXPENSES per student, not SPENDING per student. It's semantics, but I think it's an improvement because it is less likely to mislead you to think that all the money is being directed towards items that contribute to the undergraduate experience--and that more dollars will always mean more quality or more benefits for students. Lower expenses don't necessarily mean a campus is stingy or is stiffing its students. It can mean it's lucky in terms of cost of living, and/or that it has purdent practices and great efficiencies in important areas (energy star programs, pharmacy benefit managers, a purchasing consortium, you name it).</p>

<p>Naturally some of these same things apply if you use NCES data (or any other method) for finding out "who spends more" although some would drop out if you used a measure less gross than the entire operating budget.</p>

28

<p>
[quote]
Keep in mind that if the operating budget reflects deficit spending, then that rate of spending per student may not be sustainable without some kind of change in the future, either an increase in debt service or a cutting back in expenditures.

[/quote]
</p>

<p>Absolutely. That's why I strongly recommended reading the school's own management discussions and strategic planning documents. If a school is "not in equilibrium" (i.e. spending down the endowment after inflation), that will the the first paragraph in any such discussion.</p>

<p>I look at three key indicators: </p>

<p>Net student charges (after financial aid) shows what a college is really charging its average customer. </p>

<p>Per student operating expenditures shows what a college is spending for the average student's undergrad experience. </p>

<p>The third is per student endowment (and per student endowment spending). I calculate the endowment spending as a percentage of endowment value. Anything approaching 5% raises questions about sustainability based on today's conventional wisdom. Anything over 6% is a red-flag that the institution will be looking to make serious budget cuts.</p>

<p>
[quote]
Also keep in mind that sometimes half of the operating budget at these kinds of institutions go to salary and benefits. Thus, variations in cost of living and other factors can mean that an institution spends a lot less on these items while still maintaining the same number (and quality) of faculty/staff/admin per student.

[/quote]
</p>

<p>Here is where we disagree. Certainly there are significant differences in cost-of-living reflected in payrolls. Why? Because desireable locations where more people want to live have higher cost-of-living. I would argue that the desireability of a location is part of the undergrad experience. For example, take Wellesley. It is located in one of the nicest (read "most expensive") neighborhoods in the United States. If you loaded Wellesely on the back of flat-bed truck and moved it lock, stock, n' barrel to Hattiesburg, Mississippi, would it be the same "quality" school? Nope. Why? Because Hattiesburg doesn't offer what metro-Boston does. Neither students nor faculty would find it as desireable. That "desireability" is every bit as much a component of "quality" as any other factor. [Although, in fairness, there is no restaurant in Wellesley that serves a plate of bar-b-q prime rib as tasty as the Wagon Wheel in Hattiesburg!]</p>

<p>Or, to take a real-world example: Why does Grinnell have to make heavy use of merit-aid discounting to fill their class in Iowa while Haverford and Bowdoin do not? Obviously, consumers are factoring location into the "quality" equation. Seems reasonable. I would expect to pay more to live for Mainline Phila. or coastal Maine than I would in Iowa and I would perceive a quality of life component associated with those locations just like all the other people who have driven the property prices up.</p>

<p>
[quote]
I also think debt service is worth looking at, particularly in terms of what it is being spent for. If it is spent on things that students are still benefitting from, then it's a non-issue. If it's spent on things that are no longer an identifiable benefit, well, that's falsely inflating the "spending per student."

[/quote]
</p>

<p>Absolutely. Debt service is the one area of spending that gave me pause. But, in every case I've looked at, the debt service is related to specific building projects that have a definite impact on the undergrad experience. These days, it's usually a new science center, a new student center, a new athletic center, or a new dorm. These are all things that the college will certainly pitch as components of a quality undergrad experience, so I'm not sure why their cost shouldn't be factored in per student expenditures.</p>

<p>Philosophically, it's difficult to to grapple with debt service. For many schools, it is simply smart finance. The cost of borrowing money with bond issues is less than the return on endowment investments -- so "paying cash" for a new science center would be stupid. One way to account for this would be to look at the percentage of debt service as a percentage of operating expenses. If all of the schools are the same, then it's not much of an issue. Any school that stood out with abnormally low or high debt expense would be flagged.</p>

29

<p>Here's the list for FY2003. However, keep in mind the huge caveat that you really have to look at the underlying financial reports to get a clear picture. Also, you have to factor what each school will cost, both on a global per student basis and for you individually. For example, Grinnell spends less and charges less. Other schools spend less and don't charge less. Grinnell is also the most conservative spender of endowment money I've seen, spending at less than 3% in 2004 despite having the largest per student endowment of any LAC.</p>

<p>Keep in mind that Harvey Mudd really doesn't have humanities and social science departments and Claremont McKenna doesn't really have science or language departments -- instead relying on other consortium schools or joint-departments in those areas.</p>

<p>If anyone can point me to the 2004 Financial Reports for either Amherst or Pomona, I would be interested. I have not been able to find them. With those reports, I can calculate net per student charges (after financial aid) and per student endowment spending -- two fundamental items. I expected both to be a little higher on per student expenditures so I would like to look at them in more detail. It is possible that debt service on new construction is the difference. Williams and Swarthmore have both been on building binges over the last 15 years including new science centers. Swarthmore's number includes about $8 million in interest expense for bond issues related to the construction of a performing arts center, a new classroom building, a new science center, a new dorm, a new fitness/tennis center, and a major renovation project. Williams number includes interest expense on a new science center an a massive performing arts center project.</p>

<p>WELLESLEY COLLEGE $67,020
SWARTHMORE COLLEGE $65,546
WILLIAMS COLLEGE $60,957
MIDDLEBURY COLLEGE $58,022
AMHERST COLLEGE $55,421
BOWDOIN COLLEGE $54,579
SMITH COLLEGE $53,440
POMONA COLLEGE $53,286
HAVERFORD COLLEGE $50,824
HARVEY MUDD COLLEGE $50,170
WESLEYAN UNIVERSITY $48,912
HAMILTON COLLEGE $48,062
VASSAR COLLEGE $47,628
CLAREMONT MCKENNA $47,506
REED COLLEGE $47,269
BRYN MAWR COLLEGE $46,552
CARLETON COLLEGE $45,821
GRINNELL COLLEGE $45,288
COLBY COLLEGE $44,463
DAVIDSON COLLEGE $44,363
MOUNT HOLYOKE $44,229
SCRIPPS COLLEGE $42,309
BATES COLLEGE $40,290
WASHINGTON AND LEE $39,934
BARNARD COLLEGE $39,110</p>

30

<p>Interesteddad,
Since you asked ... here is an interesting link for Pomona College, but may not have all the financial data you need:
<a href="http://www.aspc.pomona.edu/archives/documents/2005-06/SizeOfCollege.doc%5B/url%5D"&gt;http://www.aspc.pomona.edu/archives/documents/2005-06/SizeOfCollege.doc&lt;/a&gt;&lt;/p>

31

<p>
[quote]
Here is where we disagree. Certainly there are significant differences in cost-of-living reflected in payrolls. Why? Because desireable locations where more people want to live have higher cost-of-living. I would argue that the desireability of a location is part of the undergrad experience.

[/quote]
</p>

<p>But higher wages and benefits aren't uniformly or perfectly correlated with desirability, for students or for the staff and faculty who they recruit, hire, and retain. And making that kind of assumption would certainly be out-of-step with a student who wanted a certain environment or geographic area. </p>

<p>I think my main point stands. You could have two schools that are both excellent and have recruited a faculty, staff, and administration that are equivalent in their quality and dedication. But if one school is in an area where living costs are less, then the part of the operating budget devoted to salary and benefits--no small proportion--will be less. This does not mean that the school is of inherently poorer quality or that the faculty is of a lower quality. I just don't know how one could asser this * regardless of whether or not you, personally, would not want to live there, or think students should not prefer to live there.* </p>

<p>Whether or not that cost-of-living correlates with things that the prospective student cares about is a different issue. At least, it ought to be.</p>

<p>It's not that I don't think you have a point in this--I just think it's not sound to fold it in with the idea of expenditure per student and assume that all the factors that make for a higher cost of living mean the school will inherently be that much better.</p>

32

<p>
[quote]
just don't know how one could assert this regardless of whether or not you, personally, would not want to live there, or think students should not prefer to live there.

[/quote]
</p>

<p>You can't use generic financial data as a substitute for individual evaluations of value. For example, take the most basic consideration. It doesn't matter one iota if a college charges the average student $25,000 for tuition, room, and board if I know they are going to charge me $40,000! The financial data (or anything other generic measure) is only useful in trying to come to grips with the nature of a particular school. At some point, you have to file that away in the knowledge bank and start asking, "so what does that mean to ME?"</p>

<p>The way to start looking at the financial data is on a macro-level. What do they charge the average student? What do they spend to provide the undergrad experience for the average student.</p>

<p>Clearly, the generic average student places value on desireable high-cost-of-living locations and is willing to acknowledge this value in dollars and cents terms. Otherwise schools in the high-cost Northeast region would have to match Grinnell's merit discounting to fill their classes. </p>

<p>So, on a macro level, there is a dollar value associated with the benefits of a desireable location, even though those benefits may be intangible (or even meaningless) for one individual student.</p>

<p>Note: BTW, I am not picking on Grinnell. It's just the perfect example -- an absolutely superb college in a location that is consistently viewed as undesireable to the elite college customer base. Grinnell knows there is a dollar-value attached to location because they have to budget the merit aid discounts to overcome their location. I think that understanding the role of location in college demand can give the high school student an advantage in the admissions game. Give serious consideration to excellent schools outside of high-demand locations -- they are easier to get into and more likely to offer merit money.</p>

33

<p>Remember that some places "superstar" faculty can be very well paid and yet be completely inaccessible to an undergrad...</p>

34

<p>You are changing my argument. You have switched the focus the cost of tuition. </p>

<p>What you have been promoting previously is looking at the gross operating expenses of an institution, dividing it by FYES, and calling it a quality measure. More spending = better quality.</p>

<p>My point ties back to the proportion of gross operating expenses related to wages, salaries, and benefits. This is no small proportion in higher education generally, and in LACs specifically. Which is why the cost of wages and benefits as they are influenced by the cost of living and other forces on wages is salient.</p>

<p>It is true that cost of living and other geographic differences can and will influence tuition and tuition discounting. But these are revenue items, not expense items. The focus of your argument, as I thought you laid it out, was not about revenues per student or the cost (sticker price or actual) to the student. It was about gross operating expenses.</p>

35

<p>
[quote]
What you have been promoting previously is looking at the gross operating expenses of an institution, dividing it by FYES, and calling it a quality measure.

[/quote]
</p>

<p>I don't recall ever calling per student spending a "quality" measure. I've refered to it as a measure of "value received".</p>

<p>Assuming two schools are equally well-managed, paying $25,000 for $65,000 worth of "stuff" is better value than paying $25,000 for $45,000 worth of "stuff". That holds true even if the additional "stuff" includes spending required for a desireable, high-cost-of-living locale.</p>

<p>I haven't proposed looking at just expenses. I've proposed looking at the difference between per student expenditures and per student revenues. That delta represents the "value" expressed in dollars and cents. There's no free lunch. Usually (but not always) colleges that spend the most per student also charge the most per student.</p>

<p>Perhaps an example: </p>

<p>Grinnell charges the average student $20,399 and spends $46,615 per student. The school is comfortably (some might say embarrassingly) in financial equilibrium, spending just 2.4% of endowment last year.</p>

<p>Oberlin charges the average student a little more ($22,239) and spends less ($40,135). It is not in financial equilibrium, with endowment spending at an unsustainable 6.1%. In fact, Oberlin's strategic planning addresses the issue, calling for an enrollment reduction (eliminating 100 financial aid students and keeping the same number of full-pay students).</p>

<p>Now, you would have to be nuts to choose between Grinnell and Oberlin solely on the basis of their financial statements. Both are excellent schools and Oberlin might well offer more value to a particular student, either because of their offerings, or style, or location, or because of an individual aid package. But, purely from a big-picture perspective, all things being equal, I'd rather pay $20k for a box of widgets that cost $47k to make than $22k for widgets that cost $40k.</p>

36

<p>I'm afraid I'm getting more confused. </p>

<p>Are you saying that your lists, published above, aren't gross operating expenses divided by FTE?</p>

<p>
[quote]
But, purely from a big-picture perspective, all things being equal, I'd rather pay $20k for a box of widgets that cost $47k to make than $22k for widgets that cost $40k.

[/quote]
</p>

<p>Well, I'm not sure everything will ever be "equal." I guess from an absolute perspective, you're right. But since things likely aren't equal (and to some degree, the equivalence is unknowable...)</p>

<p>One real problem lies if you assume the widgets cost $40K to make because the manufacturer used lower-quality iron and hired inferior workers than the manufacturer making the other widgets. What if that's not true? </p>

<p>If it turns out that the widgets cost $5000 less not because they had poorer inputs, but rather because the plant was able to negotiate better prices for the same quality iron, bring down utility costs via cogeneration, and so on....in essence, the widgets are cheaper because of factors that have nothing at all to do with quality, and everything to do with good luck and excellent management. Well, you'd still be right to go for the cheaper price, but you'd be wrong to assume your widgets are better. That's a mistake I wouldn't want justaparent to make.</p>

37

<p><a href="http://www.aspc.pomona.edu/archives/documents/2005-06/SizeOfCollege.doc%5B/url%5D"&gt;http://www.aspc.pomona.edu/archives/documents/2005-06/SizeOfCollege.doc&lt;/a&gt;&lt;/p>

<p>I kind of liked the Pomona College mini-whitepaper as an example which discusses the financial ramifications of adding more students to the Pomona operating budget. You can read the paper for details, but the results of this analysis are as follows (cut-and-pasted):</p>

<ol>
<li> Each additional student generates net revenue of $8,911 in FY06 dollars. The principal annual costs balancing the nominal tuition and fee revenue of $40,284 are financial aid ($13,340) and dormitory space ($5,348).</li>
<li> Each additional faculty member incurs an annual cost of $137,721. The principal components of this cost are salaries and benefits ($76,800) and academic/teaching space ($41,080).</li>
</ol>

<p>The ratio of these two numbers, approximately 15/1, is the number of students who would have to be added for each new faculty member for the growth to be financially neutral. This is a worst case scenario that assumes that all new costs would have to be covered out of existing operating budgets. In reality, of course, many of these costs would be covered by new gifts – both endowment and capital – that would be solicited for specific projects. The actual ratio of new students to new faculty required for financial neutrality in the operating budget would be much closer to the current student/faculty ratio of approximately 9/1.</p>

<p>Also interesting to me was the pie charts of "marginal costs of additional students 2005-2006" which breakout where the dollars are spent (including such categories for parking, study abroad, resident advisors, etc)</p>

38

<p>First rule of real estate and life: cost does not equal value.</p>

39

<p>
[quote]
Are you saying that your lists, published above, aren't gross operating expenses divided by FTE?

[/quote]
</p>

<p>Those lists are gross operating expenses divided by FTE.</p>

<p>I do not think that list provides sufficient data to provide a good feel for the nature of each school's financial underpinnings. Along with the per student expenditures, you also need to look at per student revenues net of financial aid (i.e. what they charge). Per student endowment spending and the rate of endowment spending are also key indicators.</p>

40

<p>The Pomona white paper has one glaring inconsistency. In one place, they talk about how the additional slots resulting from expansion will increase ethnic, socio-economic, and international diversity. For example, they talk about using the Questbridge firm to recruit low-income students and outreach to increase international enrollment.</p>

<p>Then, they turn around in their pro forma financial projections and assume that the incremental students will have the same average financial aid discounts as the current student body.</p>

<p>Those two assumptions are mutually exclusive. If the incremental students are weighted more towards low-income and internationals than the current student body, their financial aid costs will be considerably higher than the current average.</p>