GAZETTE: What is “long term” in this context?
VICEIRA: Well, any donor would like Harvard to be here for the next 400 years, like it has been here for the last almost 400. They want what they give to outlast a generation and help future generations.
GAZETTE: You described the endowment as “a pool of money.” Can it be used however the University decides it is needed?
VICEIRA: The assumption that it can be spent any way we want is not exactly true. When donors typically give, they do it in a restricted fashion. They say the money has to be used for a specific purpose: to fund this lab, to fund this chair, to fund this institute, and for nothing else. And [they] expect it to be wisely invested and preserved so it can fund this chair or lab or institute “into perpetuity,” which is usually the language used.
So an important fraction of the endowment is restricted (about 70 percent in Harvard’s case). There is also unrestricted giving that has fewer constraints on how it can be spent. But, nonetheless, there is an expectation that it is going to be wisely spent, and certainly not all at once.
You were asking how Harvard is special. Unlike other universities, each School has its own endowment. It’s pooled together for investment and administration purposes, but it is not a big pool of money that the University can decide to spend any way it wants.
GAZETTE: You mentioned stability over time. Can you talk about the endowment’s role in insulating a university from increases and declines in other revenue sources over time?
VICEIRA: Endowments can help in downturns to avoid drastic oscillations in your operating budgets.
The financial crisis was an example. The endowment suffered losses in line with the losses experienced by other endowments. But there were also constraints on other sources of funding: grants from the federal government, research money, and other types.
Without the endowment, the University probably would have been forced to a much more dramatic reduction in the operating budget than it was. The University acted responsibly and all Schools tried to be conservative in how they were spending. But the endowment acted as a cushion to prevent drastic increases in tuition or drastic layoffs.
It is also very important to mention that Harvard has a triple-A rating. It is very unusual to have an institution with a triple-A rating. A big part of that is because we have a healthy endowment. It allows us to access capital markets when many of those markets are closed to other institutions and corporations.
An example was the winter of 2009, when Harvard was able to issue debt in the public markets when many actors in those markets simply did not have access. We did, and part of that was the fact that, even with the losses the endowment experienced in the crisis, we still had a healthy endowment that could back up that debt.
GAZETTE: What would the difference be in interest paid on debt if we were only AA-rated?
VICEIRA: The debt spreads or interest differentials vary over time, but it could be as high as 0.75 percent. This might seem small, but in dollar terms can be quite significant. If we were rated BB, the spreads would be much larger, in the range of 2 or 3 percent. And the cost differential widens considerably in recessions. In recessions, access to debt markets is more constrained, and your rating makes a huge difference on your cost. Obviously the cost increases for everyone, but the spread of the cost differential between highly rated institutions and low- or junk-rated institutions increases.
GAZETTE: Is it safe to say that having a triple-A rating allows Harvard to do more with a fixed amount of money?
VICEIRA: Yes, at a cheaper cost.
GAZETTE: How much of the endowment is paid out to fund University expenses every year, and how is that figure arrived at?
VICEIRA: On average, between 4 and 5 percent of the endowment is distributed as payout to the Schools every year. That’s decided based on what we think is a reasonable amount that will [still] sustain the endowment’s value in the long run, after inflation, so that pool of money will be there, available to spend in perpetuity.
So, say you decide to spend 4 percent, and let’s say we expect inflation to be 2.5 percent. In order to keep the endowment at the current level into perpetuity, we have to make a long-run return on endowment of 6.5 percent. If we think of spending 5 percent, you’re talking a 7.5 percent return.
So, the question is: “What reasonable rate of return do we think we can obtain by wisely investing the endowment without incurring undue risk?” If we think that number is something like 7.5 percent, we can afford to spend 5 percent per annum on average and still maintain the real value of the endowment — the inflation-adjusted value of the endowment — for the long run.
GAZETTE: So is it a balancing act really between …
VICEIRA: It’s a balancing act between what we think we should spend — and how much we expect inflation to be — and what we think we can reasonably get in the long run without exposing the University and the endowment to undue investment risk.
Say we concluded that 7.5 or 8 percent is a reasonable return on endowment in the long run. There are not safe assets today that pay you that kind of return. In fact, the safest securities out there in the long run are inflation-protected Treasury bonds, or TIPS, and 30-year TIPS are paying something like 1 percent per annum, plus inflation. So clearly we need to take some investment risk.
A recent op-ed in The New York Times suggested higher education endowments should spend 8 percent per year. At 2.5 or 3 percent inflation, that means we should be targeting a return on the endowment of 11 percent per annum. It seems to me there’s no way in today’s capital markets that you can do that without assuming a lot of investment risk, which would expose the endowment to potentially very large losses in market downturns.
I should also mention that the type of inflation that we face as a university tends to be, on average, higher than the type of inflation reflected in the Consumer Price Index because we tend to be more consumers of services and exposed to more health care inflation. The University tends to measure inflation in something called the Higher Education Price Index, which tends to run higher than CPI inflation. So we’re not looking at just 2.5 percent CPI inflation, but probably more like 3.5 percent, which poses even more demands on the returns of the endowment.
GAZETTE: So in order to spend 8 percent of the endowment a year, you would have to add an increment for inflation, and then with that as your investment target you would need to pursue riskier investments?
VICEIRA: Much riskier investments than is currently done.
GAZETTE: Which may work in some years, but in other years ….
VICEIRA: [It] might not work, and expose us to large losses that ultimately might backfire, in the sense that they might end up depleting the endowment. While it is true that the endowment has performed very reasonably, I would not bank on the past to project the future.
The prudent thing is to be conservative about what we should expect going forward. Of course, if one has the view that the endowment should be depletive, you can spend 8, 20, 25 percent per annum and deplete the whole endowment in four years, but is that what the donors intended? No.
Could that even be feasible? Probably not, because many of these endowments are restricted on how they can be spent. But even if they weren’t, is this something that our alums would like us to do, or our donors would like us to do? Quite frankly, I don’t think so.
GAZETTE: The endowment, as you mentioned, has exceeded 8 percent returns with a fair amount of regularity, but it’s fallen short of that as well.
VICEIRA: One example is 2008-2009 fiscal year.
GAZETTE: When it lost 30 percent of its value.
VICEIRA: About 27 percent.
GAZETTE: Has the share of the operating budget paid for by the endowment been increasing?
VICEIRA: I believe the distribution of the endowment has increased well above inflation historically. Whether that becomes a larger fraction of the operating budget is a different matter. That depends on how much you decide to increase expenses at each School, and how much revenue from other sources increases. If I get a dividend increase well above inflation, but decide to increase my expenses even more, my dividend distribution now funds a smaller fraction of my budget. Historically, some Schools have kept the fraction of their operating budgets funded out of endowment distributions steady, while others have seen increments.
GAZETTE: Harvard manages its endowment using both in-house and external managers. How did that system come about, and why not just use external managers?
VICEIRA: Harvard is one of the very few that manages an important fraction of the endowment internally. That was a conscious decision because it was deemed — and I think it’s right — more cost-effective.
If those monies were managed externally, as happens at most other endowments, the amount paid in fees would wind up exceeding what we pay to the internal managers. And the compensation of these managers is based on whether they can get better performance than, say, passive investments.
So when there is a performance payment, it is because we are better off after paying the managers than we would be if we were paying for a cheaper but passive investment in that type of asset class or for an external manager. The question is: Do we get more than it costs us?
GAZETTE: How does the system of financing a private university with the assistance of an endowment compare to other models of university financing? I’m thinking of public universities that rely more heavily on money from the state every year. Is one more financially stable than the other?
VICEIRA: Public or private universities that don’t have significant endowments typically have to fund operating budgets from tuition. And, in the case of most public universities, [with] state governments allocating resources to them, that allocation is very sensitive to fiscal crises in the state, recessions in the state.
It’s not a high priority. It seems in many cases more like discretionary spending than probably it should. And it doesn’t grow very steeply over time, if at all, after adjusting for inflation. These institutions that don’t have endowments have a lot less flexibility in how they can fund capital investments necessary to maintain their physical plants, maintain their labs. They have a lot less flexibility in hiring talent.
I think that’s very important. Luis Viceira at the Business School just needs someplace to sit and a laptop or a desktop. But if you are bringing in the world expert in Alzheimer’s to work at Harvard, at the Medical School, say, or the top researcher in genetics, they typically need huge investments in labs, physical plant, lab equipment, enormous teams. And talent is very scarce. Institutions that depend very heavily on tuition revenue or public subsidies are extremely limited in what they can do.
[Faculty] need resources to be able to do their best work, and the endowment makes a big difference in having the flexibility to be able to attract the best and most talented researchers and teachers you want. It’s important.
It’s a distinctive advantage of this University, with consequences for the University and for future generations of students, both undergraduates and graduates.
By Alvin Powell, Harvard Staff Writer