The cost of running Harvard

This is from the 28 August 2017 Bloomberg Businessweek:

Salaries and wages: $1.8 billion

Services purchased (catering, security, etc.): $583 million

Benefits: $530 million, roughly equal to the revenue from graduate programs

Depreciation: $338 million

Real estate (leases, utilities, etc.): $345 million

Other (subcontractors, publishing): $323 million

Supplies and equipment run $257m, scholarships $142m, and interest on the debt $235m, with travel expenses, advertising, and postage at smaller amounts.

Total operating costs are $4.7 billion, with undergraduate tuition covering 6.4 percent of that, graduate tuition covering 11.2 percent.

That is from Kyle Stock, I cannot find it on-line.  (I am a biased source, but do note that the new, gated version of Bloomberg Businessweek is consistently excellent.)  One of the difficulties with scaling up, of course, is that Harvard cannot always so easily scale the quality and resources of its donors.  “Harvard as we know it” may be as large as the current set of donors can support.  And “Harvard as we know it” likes…”Harvard as we know it,” not some other Harvard.


"interest on the debt $235m" that is a lot of interest, they must have big debts. Why don't they pay down that debt with their endowment?

Leveraged investments, probably. Which is sound if you can get a good rate on the debt.

Could it be that they are still repaying debt taken on in the wake of the brilliant investment management by Summers?

There may be contractual reasons behind it. For example, often donations are restricted and must be spent for a specific purpose. Interest on debt usually isn't one of them.

I'm not saying they simply can't do it, just that this may require a bit more digging.

As a non-profit, Harvard can borrow at tax-exempt rates. If you can issue tax-exempt bonds, and invest in taxable bonds, and pay no income tax, that makes a lot of sense.

BTW, I am perfectly aware that there are a lot of rules designed to prevent that sort of arbitrage, but there are also ways around them, which Harvard uses.

I just finished Peter Lindert's chapter on education in the USA (TC recommended Lindert's book about 10 years ago).


1) US students are in the middle of the road among Developed Countries because: (1) minorities drag down the averages, but even adjusting for minorities, US students are just average students. Sorry but it is what it is.

2) US teachers are slightly stupider than from 50 years go, but it's slight, not a radical change. One reason is they don't get paid that well.

3) the USA favors "quantity" (no high school student is ever flunked out on purpose; though they drop out) over "quality", compared to other countries. Harvard probably bucks this trend, hence seems "elitist" but this is true in many other countries too (Tokyo university, Moscow state university, etc). Lindert doesn't get into specifics, just makes the general claim in the USA it's quantity >> quality.

Bonus trivia: due to a late 1970s California Supreme Court decision called Serrano v. Priest, 60% of funding for CA public schools comes from state revenues, not local property taxes like in many states. Proposition 13, freezing property taxes, also limits property taxes to about 25% of school funding. Federal funds constitute only 9%, and lottery/other sources are 6%.

There you have it. You lerned something reading my post. Compare to the other people that post here.

What do you mean with No. 2? They are not paid well compared to what? compared to the past? compared to other countries? compared to other administrative staff?

In general I wonder about the distribution of salaries in modern universities: Teaching Staff vs. Administration.
I cannot speak for the US but in Germany, state level administrative staff is ever increasing compared to actual teachers...

"What do you mean with No. 2? They are not paid well compared to what? compared to the past? compared to other countries? compared to other administrative staff?"

Agreed. The evidence indicates that the Total Compensation for teachers is well above median wages for the US. It's far higher than it was historically in the US. It's set at a level to easily attract bottom tier college graduates.

"I cannot speak for the US but in Germany, state level administrative staff is ever increasing compared to actual teachers"

This is certainly true in the US. Administrative positions are a far higher percentage of the teaching force than the historical average.

"US students are just average students. Sorry but it is what it is." I have taught American exchange students in British universities. They were all from good or excellent universities. All bar one were good or excellent students. But they were a year behind: in American jargon (if I've got it right) junior year students were taking British sophomore classes. What struck me was that this remained true in the course of several decades of remorseless lowering of standards in British secondary schools. What inference should I draw from that?

Taiwanese high school kids are years ahead.

And then they go to college where they goof off and end up graduating at the same basic level as Americans.

Just sayin' that being ahead in math two years doesn't really help you that much, if you end up at age 22 roughly the same level.

That said, my daughter is in very advanced math. It blows me away that kids in 6th grade can learn algebra...Maybe there is some advantage in absorbing the lessons earlier?

It blows me away that kids in 6th grade can learn algebra

Why? It's mostly pretty obvious stuff. It ain't calculus. Yeah, they're likely not going to figure out quadratic equations on their own, but the large majority of SAT level math? Easy enough to figure out

"That said, my daughter is in very advanced math. It blows me away that kids in 6th grade can learn algebra"

I took pre-Algebra in 6th grade in 1981 in middle Tennessee.

Granted, it was the hardest math class I ever took, up to and including Differential Equations.

Typical idiotic Ray post: "I read a book! It's gospel!"

@Max - I am not sure, I think compared to other people in other fields in the present; compare to Europe where teachers are better paid and students are a bit smarter (not much but a bit, as dearieme says, one year or so ahead

@dearieme - thanks for that input. I did not know you were an educator. I myself have also lectured, for free as a resume booster

@Ted Craig - trolling? You got my attention. What is your value-add here? { }. BTW I disagree with Lindert's main thesis, which is the welfare state is not so bad. The problem is, as he clearly says, it's hard to disprove. Very hard. As Lindert says, the only way you can disprove his thesis is with a hypothetical counterfactual where with less of a welfare state than in the 20th century, you would have more better output / outcomes, yet GDP growth in the 20th century, with all the wars and all the Big Government, still grew as good as or better than the more laissez faire 19th century or earlier eras (much better than earlier eras, though the Roman Empire had per capita living standards not seen until the mid-1800s) . But 'what-if' is metaphysics, not science. In a way, it's like with patent law, and AlexT's constant trolling on this theme: if more patent protection is needed, why did Einstein, Newton, etc etc etc invent what they did without any patents? If it ain't broke...

" As Lindert says, the only way you can disprove his thesis is with a hypothetical counterfactual where with less of a welfare state than in the 20th century, you would have more better output / outcomes, yet GDP growth in the 20th century,"

For example, the US having a smaller welfare state than Western Europe, but 3 decades of higher GDP growth.

"US students are in the middle of the road among Developed Countries because: (1) minorities drag down the averages, but even adjusting for minorities, US students are just average students."

Surely the effects of minority students in the US are less than the effects of minority students in other developed countries even if the proportions are higher. Black and Hispanic students in the US are orders of magnitude ahead of groups such as Pakistanis in Britain, Algerians in France, or Turks in Germany.

Proposition 13, freezing property taxes.

Proposition 13 does not freeze property taxes. Property taxes can and usually do rise 2% a year. This way people do not become unable to afford to live in their house as was common in the 70's.

and of course when a property is sold taxes rise to 1%(+bonded indebtedness)of market value.

Sorry for my total ignorance, but what does this mean? That the full 82,4% of the budget is covered by the endowment? Or there are contributions by some State (local or Federal) institutions? Or some other way to monetize the research or the brand? Because if it is not the case, to find every year 4Bn$ for free is a tough proposition indeed.

Also, it would be interesting to have a breakdown by faculty. I guess the MBA is profitable for example, and the executive one particularly so. What about, say, Medicine? Or Physics?

It's a way of whining about how expensive it is to run a school these days, most probably to rebuff the plebs' demands for higher student numbers (there was a Twitter about this vs. Canadian Unis a few days ago). But I agree this is meaningless; if they ran on the Canadians' cost strucure they would be able to put many more students through college. But that's not their purpose in this world. I always thought top US schools are about providing comfy retirement for Nobelists as an additional incentive to win the Nobel in the first place.

Donations like the $2.5M Jared Kushner's father coincidentally gave around the time his son was being considered for admission.

Must Try Harder.

"Sorry for my total ignorance, but what does this mean?"


yeah, this numeric accounting summary of Harvard finances completely ignored 80%+ of the input/income/credits side of the ledger. Odd that a professional economist did not find that omission noteworthy.

Also be interesting to see a CPA team analysis of the soundness/efficiency of Harvard financial management. Judging by the many Harvard graduates who go on to poorly manage big government budgets, it's hard to believe the Harvard financial administrators are competent -- more likely that Harvard's huge endowment/resources minimize any need for careful budgeting (US Congress operates with similar financial freedom ... and has now totally eliminated formal budgeting)

Harvard's campus is less opulent than you might think and is quite hemmed in. The main campus in Cambridge is 210 acres (85 ha).

Harvard has been expanding across the Charles River, but has been getting unsurprisingly gouged by local interest groups, Big Dig style.

In contrast, Stanford's campus is an insane 8,180 acres in the heart of Silicon Valley. It's total number of undergraduates has grown only from 6,630 in 1980 to 7031 in 2017. Over the same era, California has grown dramatically.

Stanford's endowment is also slowly but steadily catching up to Harvard's, thanks to donations from tech industry alums.

It would be reasonable for Stanford to announce that it intends to double the size of its undergrad student body over the next, say, 20 years, but there is very little political pressure on elite colleges to do this. Stanford's 12 square mile campus must be worth, I dunno, $100 billion? But Stanford pays virtually no property taxes. Yet practically nobody is pressuring them to make more use of it to educate more students.

... how did you calculate/reason the "reasonable" size of Stanford's student body ?

likewise, how big should Stanford and Harvard campuses be ?

Over an acre an undergrad is hilariously opulent.

Well, you could suggest it match the growth in California's or even the nation's population.

He's suggesting they have the space.

1 student per acre does seem very large. Did I do the math right?

It is good(?) that Leland Stanford wanted a large horse ranch.

"Stanford’s campus is an insane 8,180 acres in the heart of Silicon Valley. It’s total number of undergraduates has grown only from 6,630 in 1980 to 7031 in 2017."

Though not as selective as Stanford nor in Silicon Valley, Sewanee: The University of the South (that's its actual name) has Stanford beat when it comes to low population density. 13,000 aces -- and all for a small college with well under 2,000 students total. So over 6 acres per student. They own pretty much an entire mountain, covering so much land that forget about the Quad or the Campus -- they call it "The Domain".

There's a lot of cartel behavior among top colleges in terms of expansion of undergraduate student bodies, with all the colleges eyeing each other. Nobody is aggressively expanding, so nobody else is either. All the Harvard-Stanford-Yale-Princeton-MIT cartelistas have been content to let demand increase much faster than supply in order to become more elite.

Millennials are the first cohort to exceed boomer population. Looks like their kids will be below replacement and immigration will be down.

Full time college and university population will again shrink like in the 80s. Then, the growth in part-time community college and the decline in trade schools replaced by job focused "college degrees" masked that as the many small colleges expanded or created "night school" programs, often in partnership with States. But logistics led to separate facilities specifically for community colleges. And colleges up scaled their campuses or sold or closed or changed purpose. Those that sold to for profits 10-20 years ago are closing if not already closed in the past few years.

Of course, colleges closing in flyover, now Trumpland, would be little noticed. In many cases, town-gown hostility was high, with little in the way of jobs as the campus did no expansion or remodeling moving to deferred maintenance, so no big layoffs, nor cuts in property tax revenue.

Chicago had doubled its undergraduate enrollment in the last 30 years. Some schools don't stand with cartels.

How is that even a cartel? There's no benefit they derive from getting bigger, is there? The benefits presumably go to the students

It's almost triple the enrollment of the worst year (class of 1973).

Wrong. Yale has just added 800 students with the addition of two new residential colleges; a 15% stepwise increase in student body size.

One reason Harvard has thrived is because it realized that nickel-and-diming the parents/students who provide ~5-10% of the operating expense isn't a sensible strategy. Instead, happy memories lead to happy alumni. If the alumni also have good job prospects, they can be asked for much more money over 15x more time, particularly when they are financially more capable.

Mismanagement at elite universities leads students to count which meals to skip or meal items to forgo because the dining facility cost center is being squeezed mercilessly by middle management. A certain amount of austerity breeds esprit de corps among the students, but much of the time, it also makes them resent the administration - and that brings real costs to schools.

These statistics are about as relevant to the cost of providing a university education as the Amazon's bulk expenditures are to the cost of running an online bookseller.

Just as Amazon's revenue from selling books is small relative to total revenue, Harvard's revenue from tuition is small relative to the total. Harvard runs a very nice federal R&D contracting business, generating $500 - $600 million or so in revenue per year. They also (notoriously) run a very expensive investment management business. What fraction of the $1.8 billion in labor costs derive from the R&D or investment management operations?

One of the things I've learned from many years of reading MR is the importance of marginal costs. I'm disappointed we didn't see any discussion of the marginal costs of increasing enrollment.

Tuition at Harvard is like the fares charged by the airlines: each customer pays a different amount. Well, that's somewhat of an exaggeration, as all students from low income families (less than $65,000) pay zero. Harvard is not alone: I believe all elite colleges charge different tuition to different students. And the differences aren't based solely on income, as the elite colleges offer scholarships to students from high income families as a recruitment tool. Elite colleges have very large endowments (Harvard's endowment is almost $35 billion), so they can be as flexible with tuition as they wish. As they wish. Some make a persuasive case that the elite colleges with their large endowments aren't really colleges but big businesses (investment bank, R&D, etc.) that offer a few classes to maintain the fiction of their tax exempt status. Of course, the elite colleges could alter the tuition structure. For example, they could charge tuition based on the incomes of their graduates, collecting zero while they are students and a percentage of the graduates' income for a defined period after graduation. The point is that the elites, being mostly private, have flexibility to adopt almost any model they wish. Elite public universities, however, don't have that flexibility, even those with large endowments. Indeed, elite public universities are under constant political pressure to reduce tuition and spend the endowment instead. This combined with the mandate that a minimum percentage of incoming freshman must be residents of the state (which dilutes the quality of the students). In the competition for the best students, the private elites have an enormous advantage, so cry for the public elite university not for Harvard.

My understanding is that the Ivies do not offer merit scholarships. You have to drop down a level (e.g., to Vanderbilt) to see those.

as the elite colleges offer scholarships to students from high income families as a recruitment tool.

Perhaps if the high income family submits its financial information to the University for its perusal. Many high income families would rather pay the tuition.

"And the differences aren’t based solely on income, as the elite colleges offer scholarships to students from high income families as a recruitment tool." Are you sure about this? I thought (and google seems to agree) that the Ivy Leagues don't offer any academic scholarships at all.

Why on earth do people donate money to their universities? Unless it's a huge amount, it doesn't even get your kid in. Aren't there better causes out there?

Lame. See above: must try harder.

Berkeley's budget:

I believe its a much bigger school with a much smaller budget.

Note the huge amount of Harvard's budget that is fedreally funded:

Only a fraction of Harvard's efforts are devoted to educating students. Maybe less than half I would guess...

For all you Harvard grads,

Here is a tough math question:

How many Jared Kushner's

Would Harvard have to admit

To pay for faculty and other salaries.

No cheating.

Are Harvard's financial statements presented in accordance with GAAP?

Only an academic could "crunch" the few numbers provided in the post. Expenses have little meaning without quantifying "revenues." To wit, salaries and benefits should be added.

A financial analysts would be concerned with levels and trends. Three or four simple Lotus spread sheets, common-sized and growth-comparative statements would start the analysis.

Anyhow, the buildings and land; the entire endowment (in brown paper bags filled with small-denomination, unmarked bills); the motor vehicles; etc. should be handed over to black and brown people. Black Lives Matter!

I was simply paraphrasing a BLM leader.

But, Kudos!

You're an iota more discerning than the typical liberal idiot (redundant); or the common Stalinist, black supremacist, black nationalist, American-hating terrorist.

Anyhow, you responded, Dog.

Actually, I'm more of a unicorn.

Here's a challenge: Give me the quote of the BLM leader that that said "the motor vehicles; etc. should be handed over to the black and brown people" along with the link.

Or, was it in your mind.

This one's for you Bill-

D the B knows of which he speaks.

You'd think that would have gotten more coverage nationally.

As much as some neo-nazi's rantings.

I'd love to see the media ask Democratic candidates about this idea. Do they disavow it or support it?

A fair media would make them go on the record.

Also, they need to respond immediately. No waiting a few days.

Neither Rich nor Harun have provided evidence for the claim: "the motor vehicles, etc; should be handed over to the black and brown people..."

So, my car is safe.

Sorry, Bill, when the ship is sinking, even the rats have the good sense to leave.

Abandon ship!

Rich, You're belief system is not a life raft, it's a rock. I can't save you from what you want to believe, but understand most rational people don't buy what is being offered for sale.

Have you considered the logistics of this. If each bag contained $10,000 there would be 4 million of them. And the security !

Yes, it's standard procedure for all colleges to annually issue their audited financial statements, using FASB if they're private and GASB if they're public. Harvard's fiscal year 2016-17 numbers are probably still being audited, but their 2015-16 audited financial statements are available here:

In addition to Harvard's balance sheet, it has an income-and-loss statement where you can see where its revenue came from and where it spent the money. Endowment income is its largest single source of revenue*; another sizable one that hasn't been mentioned by commenters here is donations.

*"Revenue" should be in quotes because it's really the endowment income that was made available for operations (instead of being re-invested).

P.S. Most colleges and universities make their financial statements available to the public, but if you want to quickly compare the finances of a lot of colleges, it's faster to download the data from the IPEDS website that I mentioned in another thread.

In the United States, education, as everything else, is at the paws of malefactor of great wealth, remorseless plutocrats who prey the populace.

Approximately 22,000 students in total, so it's costing over $200,000 per student. Seems a bit high.

Harvard does a lot more than educate students. The professors produce a lot of research, and the numbers may also include the hospitals (not sure).

Assuming Harvard roughly breaks even from year to year, does the rest of the revenue come from profits earned on the endowment?

Assuming revenue = expenses ($4.7B), then that means the 6.4% raised from undergrad tuition amounts to $300M and the 11.2% raised from graduate tuition amounts to $526M.

If I were Harvard, I would raise undergrad tuition by 33% to $400M and raise undergrad need-based scholarships by a corresponding $100M.

Rationale: You're Harvard. Only a handful of U.S. universities (Stanford, MIT, Princeton) can credibly claim to rival you in terms of your degree's "signalling strength". Ergo people should be willing to pay a "premium" for that degree. Only, you don't want to completely price the no-wealthy out of the market. So, generous need-based aid.

Harvard (and some other places) already do this to a large degree; it's completely free for students whose household income is less than $60k to attend Harvard. I used their online calculator and compared how much I'd pay at Harvard vs. the best "large state school" in my state, and it was only marginally more. My household income is around the 10th percentile nationally.

It doesn't make sense, though, for Harvard to charge a top 0.1% household roughly the same amount that household would pay to send a kid to Cornell, USC, Tufts or Boston College. They're giving up revenue needlessly.

It doesn’t make sense, though, for Harvard to charge a top 0.1% household roughly the same amount that household would pay to send a kid to Cornell, USC, Tufts or Boston College. They’re giving up revenue needlessly.

Oh, they expect to get money from them in non-tuition contributions. No need to piss them off by jacking up the prices they're required to pay.

Note that scholarships are in the Kyle Stock list of of operating costs at $142 million. Scholarships are, however, just a discount to revenue, and are not an expense. It is sort of like calling the Standard Deduction for U.S. personal taxes an operating expense of the federal government, when it in fact just lowers revenue to the U.S. government relative to stated tax rates.

Of the 100 best American poets born 1900-1950, exactly one went to Harvard. Which means it is just as much a provincial place in the poetry world as, say, SUNY Geneseo. My guess is that Harvard grads also run at about 1 in 100 of the best recreational mathematicians of that cohort, 1 in 100 of the best chamber musicians, and 1 in 1,000 at best of the best actual mathematicians (spare me your criticisms - I actually know what I am talking about on those subjects - please note I said best of the best, not the really really good. Sample size is a little small of course, so I could be off by orders of magnitude. I probably am not, but I could be.) Not a provincial wasteland by any means but not much more than a place where lots of people who are good at taking tests congregate for a few years before moving on to better pastures. Still, they probably can take credit for 20 of the best 100 classicists, which is why I feel enamored of the place, to a certain point. That being said, while the professors are pretty much flashy mediocrities (with a sprinkling of Aspergerian heroes and rabbinically insightful moral geniuses ) (ratio of five to one in favor of the poor Aspergians - you can look it up) the students, almost all of whom are young, healthy, and opinionated, must have really interesting conversations. If I had big bucks I would take that movie about Harvard - Social Network - and turn it into a multi-year multi-episode series, focusing on the late night conversations of youth and the subsequent disappointments and achievements - sort of like Balzac's Grandeurs et Miseres des Courtisans, but with really good New England cinematography. And when you think about it, really good New England cinematography is gold, pure gold, as Seinfeld said to the Carrot Top stand-in (actually it was vice versa - Jerry, that is pure gold: those are the five words you want go google if you don't believe me).

Update and correction: Kenny Bania: That's Gold, Jerry, Gold! (Jerry went to SUNY Oswego, not Geneseo.)

another update - if I had big bucks - nah, I wouldn't waste my time turning movies into mini-series. For the record. I'd spend as much of it as I could in Rhode Island, helping people who will never be filmed. Why Rhode Island? Why not?

For the record, I am absolutely fascinated by New Jersey - almost as much as Tolkien was fascinated by Middle Earth, or would have been, if it were, like, real: but Rhode Island is also extremely fascinating. Had Tolkien been born in America, he would have gone to Harvard, so they would have 2 out of a 100 of the best poets of that cohort (well if Tolkien were young enough). You know how when you watch a Western and there are a bunch of guys riding horses in a valley underneath mountains that have been there for so many many millions of years and they act as if they are going to 'tame that valley'? Ridiculous. Mock New Jersey all you want, or Rhode Island, but you are never going to tame that valley, either. God loves us the way we are but loves us too much to let us stay that way. Providence, Newport, all that water, all those boats. The rain that day, when such nice promises were made. And, all these years later, I can say - they were kept! That, my friends, is not ridiculous.

thanks for not criticizing. I know how few people tried to read this with understanding, and, for the record, I don't care (I also have stopped checking back to see criticisms of my short prose poems on websites, although I remain grateful for the bandwidth,,,,). It is no easy task to try to teach bots (as we now call them, they will have better names in the future) to talk, which is one of the things I am trying to do. Kindness, for them, is more important than philosophy. Philosophy is all well and good for most of us: but we are elites in so many ways and elites can get addicted to philosophy at the expense of kindness. Kindness, for them, is more important than philosophy. Wallace Stevens once said that no evening reading of novels could replace, for him, the opportunity cost of walking down the evening streets of New Haven. If you do criticize, though, remember this: I have dozens of elderly nuns, and dozens of other people who might as well be elderly wonderful nuns, on spiritual speed dial - and those who criticize me fairly, as well as those who criticize me unfairly, are equally benefitted by my requests to those nuns to pray for the people who care enough to criticize, fairly or unfairly. But to get the benefit of those nuns praying for you, there is no need to criticize: consider it done! The speed the speed of it all: nee lyog-kovo Rabawta eez Balawta taskatzs BeegeeMawta. (Russian children's rhyme, roughly translated: it is no easy task to rescue a hippopotamus from a swamp). And yes, I can see why someone might think Harvard is more interesting than Rhode Island or even my beloved New Jersey. God loves us all, I understand that too.

for the record: yes, the promises were kept. Both von Neumann and Stevens understood that in their last days of life (lots of good things happen that started in New Jersey and New Haven). For the record: yes, the promises were kept.

This sounds a lot like modern football (soccer) economics.

are salaries at Harvard too high ?

The Faculty of Arts & Sciences budget was $1.3 billion in 2015.

The FAS budget is both large (approximately
$1.3 billion) and highly decentralized,
with significant spending under the direct
control of over 150 separate departments,
centers, libraries, and museums. The
Consolidated Statement of Activity presents
important categories of revenues and
expenses of the FAS as a whole. This view
combines what is typically called the
“Core” of the FAS (which comprises the
faculty, the College, and the Graduate
School of Arts and Sciences) together with
the other major affiliates of the FAS
(i.e., Athletics, the Division of Continuing
Education, Dumbarton Oaks, the
Harvard College Library, the Museums,
and the School of Engineering and
Applied Sciences). Given that the Core
constitutes about 73% of both the
FAS Fiscal Year 2015 consolidated revenues
and consolidated expenses, we also
present a Fiscal Year 2015 Statement of
Activity for just the Core.

Comments for this post are closed