Harvard's Annual Financial Report Fully Details 2009 Losses

Updated

The Harvard University Financial Report for fiscal year 2009, published October 16, contains more than the usual amount of dramatic material, headlined by the $11-billion loss of endowment wealth—the most important factor driving budget reductions throughout the institution. The report’s texts, footnotes, and tables merit especially attentive reading for the insights they offer into

• the full (and continuing) losses, which may ultimately exceed $1 billion, on Harvard’s interest-rate swaps associated with its borrowings;

• the dimensions of the losses incurred in the “general operating account” (GOA)—the principal funding mechanism for University operations (where the assets, invested like the endowment’s, absorbed proportional losses plus the swap-related costs); and

• further details on the performance of the investments, overseen by Harvard Management Company (HMC).

First, an overview of the year as a whole.

The Big Picture

“Notwithstanding the challenges we have faced during fiscal 2009,” writes Daniel S. Shore, vice president for finance and chief financial officer, and James F. Rothenberg, Treasurer, “Harvard’s financial foundation is strong and will continue to enable the University to deliver on its guiding purposes: to achieve excellence in research and education; to prepare students for leadership and for lives of meaning and value; to advance the course of knowledge and ideas; and to serve society.”

In a conversation, Shore acknowledged the obvious: that it was “a challenging year for us.” Once the challenges became clear, he said, the University set about managing through the problems to prepare to adapt Harvard to what the report calls “a new economic footing” for the longer term. He pointed to the strength of Harvard’s faculty members, students, facilities, and long-term potential for growth in Allston, and its continuing relative strength among universities, in spite of an adverse period during which the institution “certainly lost significant wealth.”

Evidence of that adaptation, and the relative strengths reflected in the scale of Harvard’s operations, emerge from the financial statements. For the fiscal year ended last June 30, the University in fact achieved an operating surplus of $71 million, up from a $17-million surplus in the prior year.

That result reflects revenue growth budgeted before the crisis in the financial markets and the ensuing recession, followed by internal efforts to cut spending progressively as the dimensions of the emerging problem became clearer.

Harvard’s revenues grew a vigorous $345 million, or nearly 10 percent, to $3.83 billion. (Fiscal year 2008 results were reported in the November-December 2008 Harvard Magazine.) This rate of revenue growth actually accelerated from the prior year. In both years, distributions from the endowment (valued at $36.9 billion at the end of fiscal year 2008, versus $26 billion now) were the driving factor. Funds from the endowment distributed to support University operations increased $241 million, or a robust 20 percent, to $1.44 billion.

(The half-percent administrative assessment for the “strategic infrastructure fund,” used to defray Allston-related development expenses, rose modestly, to $176 million, bringing the aggregate payout rate to 4.6 percent, down from 4.8 percent in the prior year, when there were also significant net “decapitalizations” of endowment funds; there were negligible decapitalizations in fiscal year 2009.)

Those trends, fun while they lasted, will now reverse: in the current fiscal year, the operating distribution is forecast to decrease by 8 percent, or more than $100 million; and in fiscal year 2011, the distribution is likely to decrease a further 12 percent from the now-reduced level, or an additional $150 million or so.

The implications are obvious: in the year just ended, distributions from the endowment accounted for nearly 38 percent of University operating revenue—four percentage points more than in the prior year. During fiscal year 2009, the Radcliffe Institute derived 87 percent of revenue from endowment distributions, the Divinity School 73 percent, and the Faculty of Arts and Sciences 55 percent, with other units scaling on down to the School of Public Health (15 percent). Every unit but two (the Graduate School of Education and the School of Engineering and Applied Sciences) depended more heavily on endowment income in the year just past than in 2008.

Other sources of revenue were mixed. Support for sponsored research rose to $714 million from $668 million (about 7 percent), led by non-federal sources. The five-year federally funded Harvard Catalyst program for clinical and translational biomedical research activities yielded $21 million during the fiscal year; a $21-million payment from the larger, private pledge for the Wyss Institute for Biologically Inspired Engineering was also received.

But revenue from students declined 1 percent, to $678 million, as higher tuition and fees were more than offset by a 20 percent increase in scholarships applied against such income.

Current-use giving (including the Wyss funds) rose 23 percent, to $291 million; but giving overall declined by $93 million, to $597 million, as gifts for endowment funds plunged $142 million (42 percent).

Expenses grew $291 million, or 8.4 percent, to $3.76 billion: not austerity, it would seem. Salaries, wages, and benefits—49 percent of total expenses—increased 11 percent, to $1.84 billion. But included in that total is $59 million in one-time severance and benefit costs associated with the staff early-retirement-incentive program and the subsequent layoffs, which resulted in the departures of more than 800 employees last spring. Adjusting for those costs, it appears that compensation expenses were still up more than 7 percent. In part, that probably reflects hiring associated with sponsored research. But it still points out the pressure to maintain controls on filling open positions, to restrain faculty appointments, and to consider whether to impose the salary freeze for faculty and non-union staff beyond the current year. The reclassification of certain rental payments also boosted reported expenses.

The footnotes tell another story about discretionary “other expenses”: purchased services (from consultants to janitors and security guards), travel, publishing, postage, telephone, and other charges declined during the year, as belts were tightened. Subcontracted expenses under sponsored projects, a large budget item, increased nearly $30 million, consistent with the expansion in such programs. “We got a good, honest start” on reining in such costs, Shore said.

The stand-out expense item, unfortunately, is trending sharply upward. According to the report, the University incurred about $58 million in increased interest costs. That figure reflects the issuance of nearly $1 billion of fixed-rate tax-exempt bonds during the year, with an effective annual interest rate of 5.4 percent (well above the cost on the daily and weekly variable-rate notes paid off in part with the proceeds), and of $1.5 billion of taxable bonds at a 5.8 percent rate. Because those debt offerings were on the books for only about half of fiscal year 2009, it is safe to estimate that the added interest expense will rise by an additional amount of the same magnitude—another $50 million to $60 million perhaps—this year.

Capital Spending

Capital spending and property acquisitions totaled $644 million, up about $50 million from fiscal year 2008. Major projects included the Law School’s Northwest Corner complex, where steel was being erected this autumn; the prospective renovation and expansion of the Fogg Art Museum; and the Allston science complex, where work will be reassessed at the end of this calendar year.

Shore said that Harvard was “still in the process of planning and thinking about the options for all” major construction projects, including the art museum; design details, construction costs, and financing are still being reevaluated.

An unanticipated but significant project is the renovation of Cambridge laboratories (and relocation of some existing laboratories) to accommodate stem-cell scientists, and similar work in the Longwood Medical Area, for bioengineering researchers; both had been scheduled to go into the Allston complex, the completion of which is delayed at best. This previously unscheduled work will be paid for as part of the Allston-related infrastructure fund, given the necessary redeployments.

Apart from those four projects, no significant work is in Harvard’s pipeline. Additional work on Allston will obviously be a long-term project. In the interim, Shore noted, Harvard has to identify appropriate uses for properties it has acquired but now will not occupy or redevelop for an extended period. Institutional uses, codevelopment options, and private alternatives for use of the land by other investors are all options to consider. “We haven’t ruled out any option,” Shore said.

Swap Losses

The report refers delicately to “realized and unrealized losses on interest rate exchange agreements held by the University as part of the financing strategy for its capital program.” As reported previously, the “notional” value of such swap agreements soared from $1.4 billion to $3.7 billion during fiscal year 2005, when Harvard put in place forward interest-rate agreements to finance then-anticipated rapid campus construction in Allston. The fair value the University would have paid to terminate those agreements, a volatile sum related to market interest rates, ballooned to $330 million at the end of fiscal year 2008.

In the chaotic conditions that shook world financial markets last fall, problems arose in refunding very short-term debt instruments, and central banks pushed interest rates to record lows. That put Harvard in a double bind: to refinance its borrowings, and to cover rising obligations under the swap agreements.

As the new report notes, the “unprecedented” fall in interest rates caused the University’s swap agreements “to incur sudden and precipitous declines in value, which in turn led to significant increases in associated collateral pledged to counterparties, creating liquidity pressures on the University.”

In response, the University terminated such agreements with a notional value of $1.138 billion during the year, buying its way out with cash payments of approximately $500 million. (John Lauerman and Michael McDonald of Bloomberg detailed some of the swap transactions; noted that other institutions—Yale, Georgetown, and Rockefeller University among them—had suffered swap losses; and quoted a municipal-finance expert to the effect that the Harvard termination payment was within the range of such costs for such transactions, suggesting that institutions limit their exposure to these instruments.)

But in addition, Harvard entered into new swap agreements with a notional value of $764 million, structured so as to offset other, existing swap agreements, locking in unrealized losses of a further $425 million. (Technically, these “offsetting” swaps have Harvard paying a variable rate of interest and receiving payments at a fixed rate; the original transactions had Harvard paying a fixed rate of interest and receiving payments at a variable rate.) This is, in effect, a financing transaction, locking in losses which will have to be realized in the future, but immunizing Harvard today from still steeper losses should the interest-rate environment remain adverse relative to the assumptions under which the original swaps were structured.

And finally, the University remains open on risks amounting to an additional $250 million of swap-related loss, not hedged by any offsetting transactions, as of the end of the fiscal year last June. Thus, during the year, it realized and paid for a half-billion-dollars worth of swap-related losses, and ended the year with about $675 million of unrealized losses remaining (the fair value of the swap portfolio, with a notional value of $3.14 billion): about $425 million locked in by offsetting swaps, and $250 million of remaining exposure subject to the market.

(For perspective on on the University’s financial and operating leverage generally, see the Harvard Magazine comment, “Liquidity and Leverage.”)

General Operating Account

Attention has understandably focused on the deflation of Harvard’s endowment, from $36.9 billion at the end of fiscal year 2008 to $26 billion this past June: it is, as noted, a hugely important source of operating revenue for the University (and the principal engine of revenue growth in recent years), and as the largest such base of assets, it bears ready comparisons to peer institutions’ performance.

But the report also draws attention to losses in the “University’s portfolio of pooled cash balances” or General Operating Account (GOA)—in effect, the funds used to pay the bills. This pool of cash, held centrally, receives, manages, and disburses cash balances held by the schools, centers, and administration. But many of the assets have been invested alongside the endowment in what is called the General Investment Account (GIA)—generating excess returns over money-market instruments, to what has been the long-term benefit of the contributors and users of the funds, but running the risk that the assets will be affected adversely in unfavorable or illiquid investment markets. Both of the latter considerations obviously came significantly into play during the past fiscal year.

Although a decision was made in 2008 to “reduce the risk profile of the University’s pooled cash investments,” and implementation had begun, the chaos that erupted in the fall of 2008 disrupted that transition, and made it impossible to shield the GOA.

The net result of a $2.8 billion decline in the value of investments, the payments on the swap losses, the infusion of the remaining proceeds from the new debt offerings, and other fund flows was to reduce the GOA’s net asset balance to $3.7 billion at the end of fiscal year from $6.6 billion at the end of the prior year. (In essence, the $11-billion decline in the value of the endowment is only part of the story; the value of Harvard’s net assets overall declined from $44.2 billion to $30.1 billion, with the endowment and GOA accounting for most of the damage.)

In an October 17 Boston Globe story (“Harvard admits to $1.8 b gaffe in cash holdings”), reporter Beth Healy quoted a statement from James Rothenberg, the treasurer, to the effect that responsibility for the investment decisions and resulting losses in the GOA did not “sit with a single individual: the Corporation plays a role, the University’s financial team, including the CFO, play a role, and I play a role as treasurer.”

Given the diverse sources and uses of GOA funds, that decline does not translate directly into reduced income streams of the same magnitude; but it does imply reduced wealth and future income, just as the loss of endowment value implies future financial constraint.

In an interview with the University news office posted October 16, Rothenberg and Shore discussed both the swap transactions and the GOA losses. “All of these losses were a function of last year’s extraordinary market conditions,” according to Rothenberg. Asked whether “the University’s investment strategies square with its responsibility to steward endowment funds,” Shore said, “There does need to be a balance between investing for long-term returns and managing for near-term needs, and we are now more conscious than ever of that balance….” He said the University recognizes “the need to match near-term liabilities with more liquid assets.” As for the large magnitude of the swaps, Rothenberg said. “Compared to most universities, our use of interest-rate swaps was certainly larger because the projected capital program that we were looking at was larger,” given the planned construction in Allston, which was “a major focus, and we were planning that expansion aggressively.” Asked whether the transitions in University and Harvard Management Company leadership exacerbated the financial-management problems, Rothenberg said, “In times of change, there are always going to be challenges, especially when you are dealing with complicated investments. With that said, no one could have foreseen the past year’s unprecedented declines in value…exacerbated by significant liquidity constraints.…What we have experienced financially is a product of these sudden and precipitous changes more than anything else.”

Responding to a query about “the Corporation’s responsibility for those investment decisions” during the transition period, Rothenberg said, “The President and Fellows have ultimate fiduciary responsibility for the University, including its finances. We take that responsibility very seriously, and we devote quite a lot of our time, especially these days, to matters of financial strategy and planning, thinking about how to balance present and future needs.” Direct investment management, he noted is conducted by Harvard Management Company, whose board he chairs. “There weren’t any reliable predictors of precisely when and how a global economic crisis would unfold,” Rothenberg said, “and there were valid arguments for why the strategies in place made sense both when they were made and right up until last fall.” In the future, he said, “I think the likelihood is that the University will continue to invest portions of pooled cash alongside, the endowment, but likely not to the same degree.”

The Endowment

The aggregate and by-line results (a negative 27.3 percent return on investments) are as reported on September 10. But a few additional details emerge from the annual report format. First, HMC president and CEO Jane Mendillo’s report reveals that the real-estate portfolio (a part of what HMC calls “real assets”) suffered a loss of “over 50 percent during the year,” after all market valuation adjustments were made. This seems consistent with and confirms general news accounts of the unfolding, very large losses in commercial real estate, both property values and loan portfolios.

Second, it would appear that foreign-exchange trading and investments—a new field for HMC, begun under Mendillo’s predecessor—have largely been wound down. Long and short positions reported at $17.5 billion each in the 2008 report had been whittled down to one-eighth that size or less as of June 30, 2009.

Third, the new footnote 4, on the fair value of investment assets and liabilities—required for the first time under Statement of Financial Accounting Standards 157—provides a snapshot of major categories of asset values, and of changes in holdings during the year, for a substantial portion of the assets, notably those that are least subject to ready market pricing.

“Active management was essential throughout this period,” Mendillo reported. “We worked decisively to make changes to our asset allocation and to increase our flexibility early in the fiscal year.” Indeed so. For example, HMC was a net seller, during the year, of $326 million of domestic common stocks; of $484 million of domestic fixed-income instruments; and of $618 million of inflation-indexed bonds (nearly all of the assets in that category, which nominally are intended to be 5 percent of the policy portfolio that guides investment holdings in the GIA). Those trades suggest efforts to make the portfolio more liquid, as well as whatever valuation decisions the portfolio managers were making. HMC was also a net seller of $1.9 billion of assets in absolute return and special situation funds—its name for hedge funds, and the largest sales, in dollar terms, in the whole portfolio (the policy portfolio allocation for such investments has been reduced for the current year).

The annual report format also gives more detailed asset categories than HMC is now using to report its results, and makes it possible to attach approximate dollar figures to the percentage investment declines HMC has reported. The rise in unrealized losses during the year—totaling some $8 billion—is concentrated in HMC’s largest and least liquid holdings: hedge funds ($1.2 billion rise in unrealized losses), private equities ($1.9 billon), and real assets, including real estate, commodities, and timber and agricultural land ($2.4 billion).

In Prospect

Shore did not forecast the University’s running rate of expenses for the current fiscal year, nor the likely change from 2009. Harvard has set its endowment distribution for the year, at a level lower than in 2009, but other factors—success in attracting more sponsored support for research, the volume of giving—will still affect revenues and expenses overall. Most schools have reserves (see the discussion of the Faculty of Arts and Sciences, for example), and are drawing them down in a planned way to buffer the reduction in endowment distributions, and therefore to lessen the cut in expenses they would otherwise incur.

He also pointed to longer-term opportunities for administrative savings, in fields ranging from procurement to the provision of human-resources expertise to information systems and technology. The goal, he said, is not to centralize any activity per se, but to find the best performers and practices, to adopt “different aggregations of activities,” and to realize productive economies across the institution as warranted. Such savings, he said, are intended to translate directly into preserving junior-faculty slots, for example—the resources and activities at the core of Harvard’s academic mission. 

From the financial manager’s perspective, he said, the new reality is maintaining a much more flexible posture toward plans and budgets, testing diverse scenarios at different levels of revenues, and helping the whole community cope with a great deal more uncertainty by assuring that the institution can be kept appropriately nimble and responsive. A Financial Management Committee, with expanded representation tapping alumni and faculty expertise, and including both Mendillo and Treasurer James Rothenberg, is better integrating University and Harvard Management Company perspectives on risk, risk management liquidity, investment opportunities, and more. It is advising Shore himself, Katie Lapp (the newly arrived executive vice president), and through them, President Drew Faust and the Harvard Corporation, where the University’s financial policies and endowment distributions are finally vetted and approved.

Within that framework, Shore said, budget guidance for fiscal years 2012 and beyond has not yet been promulgated. During 2009, he said, in light of volatile economic and financial conditions, the Corporation deferred making final budget guidance until last spring—appropriate in the circumstances, but a budget-making challenge for the institution. He said he was uncertain whether the same practice would be followed this academic year, or whether the Corporation would feel comfortable reverting to its traditional timing, providing guidance on subsequent years’ endowment distributions and budgets in late autumn.

In any event, he said, the critical balance remained the same: not cutting budgets so deeply now that essential activities were irreparably harmed, but not treading so lightly today that the cutting would have to extend many years into the future, to restore distributions from the now-reduced endowment to a sustainable level. If the balance can be set properly, he said, Harvard will sooner find itself in the happier position of being able to grow the endowment distribution again, in support of essential academic work and innovations, after investment returns strengthen.