Yale university investments office august 2006

9-807-073

R E V : M A R C H 1 8 , 2 0 1 1

________________________________________________________________________________________________________________ Professor Josh Lerner prepared this case. Josh Lerner has endowed a research fund at Yale jointly with a member of the Yale University Investments Office. HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorsements, sources of primary data, or illustrations of effective or ineffective management. Copyright © 2007, 2011 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545- 7685, write Harvard Business School Publishing, Boston, MA 02163, or go to http://www.hbsp.harvard.edu. No part of this publication may be reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means—electronic, mechanical, photocopying, recording, or otherwise—without the permission of Harvard Business School.

J O S H L E R N E R

Yale University Investments Office: August 2006

David Swensen slowly crossed the trading floor of the Yale Investments Office and looked over the hectic scene. While Swensen himself could not move quickly—he had been on crutches since an injury in the Yale Summer League championship softball game against the Biological and Biomedical Sciences team earlier in the month—the remainder of the staff was moving rapidly, whether leafing through online data or consulting with their peers about prospective investments.

Swensen had every reason to feel content, despite his recent injury. The endowment had just completed another spectacular year, having grown to $18 billion (up from $1 billion when he had taken over the office). Yale had developed a rather different approach to endowment management, including substantial investments in less efficient equity markets such as private equity (venture capital and buyouts), real assets (real estate, timber, oil and gas), and “absolute-return” investing. This approach had generated successful, indeed enviable, returns. Swensen and his staff were proud of the record that they had compiled and believed that Yale should probably focus even more of its efforts and assets in these less efficient markets.

But his thoughts turned to the larger challenges associated with the management of the university’s endowment. The very success of their strategy had generated new questions. How far did they think Yale should or could go in this direction? How should they respond to the growing popularity of the approach they had chosen? Given the turbulent times that private equity funds were facing, should this asset class continue to play an integral role in Yale’s portfolio?

Background1

Ten Connecticut clergymen established Yale in 1701. Over its first century, the college relied on the generosity of the Connecticut General Assembly, which provided more than half of its funding. The creation of a formal endowment for Yale was triggered by the 1818 disestablishment of Congregationalism as Connecticut’s state religion. Students and alumni alike demanded that the school respond by establishing a divinity school to offer theological instruction. To fund this effort, numerous alumni made large gifts, the first in a series of successful fund drives. While Yale used many of these donations to buy land and construct buildings, other funds were invested in corporate and railroad bonds, as well as equities. By the century’s end, the endowment had reached $5 million. 1 This section is based on Brooks Mather Kelley, Yale: A History (New Haven: Yale University Press, 1974); David F. Swensen, Pioneering Portfolio Management: An Unconventional Approach to Investment Management (New York: Free Press, 2000); and Yale University Investments Office, The Yale Endowment (New Haven: Yale University, 2006).

This document is authorized for use only by Man Lam (karmenmlam@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

807-073 Yale University Investments Office: August 2006

2

The growth of the endowment rapidly accelerated during the first three decades of the twentieth century. This was due both to several enormous bequests and to aggressive investments in equities, which made up well over half the endowment’s portfolio during the Roaring Twenties. In 1930, equities represented 42% of the Yale endowment; the average university had only 11%.2 Yale avoided severe erosion of its endowment during the Great Depression in the 1930s, however, because many quite recent bequests were kept in cash or Treasuries rather than being invested in equities.

In the late 1930s, Treasurer Laurence Tighe decided that the share of equities in Yale’s portfolio should be dramatically reduced. Tighe argued that higher taxes were likely to expropriate any corporate profits that equity holders would otherwise receive even if a recovery were to occur. He argued that bonds would consequently perform better than stocks. His decision, which stipulated that at least two dollars would be held in fixed-income instruments for every dollar of equity, set the template for Yale’s asset allocation over the next three decades. The treasurer and trustees continued to manage the endowment themselves during this period, selecting individual bonds and high-yield or income-oriented stocks for the portfolio. These policies seemed very prudent in the late 1930s and 1940s. But unfortunately, they were less well suited for the bull market of the 1950s and 1960s. In the mid and late 1960s, in response, the endowment’s trustees decided upon two substantial policy shifts.

First, the trustees decided to increase substantially the university’s exposure to equity investments. In this decision, they were influenced by a task force sponsored by McGeorge Bundy, president of the Ford Foundation. This committee—which included Kingman Brewster, president of Yale—argued that most university endowments had taken too conservative an approach: “It is our conclusion that past thinking by many endowment managers has been overly influenced by fear of another major crash. Although nobody can ever be certain what the future may bring, we do not think that a long-term policy founded on such fear can survive dispassionate analysis.”3

Second, Yale decided to contract out much of the portfolio management function to an external adviser. The school helped to found a new Boston-based money manager, Endowment Management and Research Corporation (EM&R), whose principals were well-known, successful growth-stock investors recruited from other Boston money management firms. The plan was that EM&R would function as a quasi-independent external firm and would be free to recruit additional clients. At the same time, Yale would be its largest client and would have priority over other clients.

The high expectations for EM&R were never realized. Like other universities, Yale saw its endowment’s value plummet in the ensuing years because of a bear market, accelerating inflation, and operating deficits. Between 1969 and 1979, the inflation-adjusted value of Yale’s endowment declined by 46%. While the investment performance was not that unusual relative to other endowments, it nonetheless severely strained the financial fabric of the university. Yale terminated its relationship with EM&R in 1979 and embarked upon a program to use a variety of external advisers in its evolving asset management framework.

2 General university information is from Institutional Department, Scudder, Stevens & Clark, Survey of University and College Endowment Funds (New York: Scudder, Stevens & Clark, 1947). 3 Advisery Committee on Endowment Management, Managing Educational Endowments: Report to the Ford Foundation (New York: Ford Foundation, 1969).

This document is authorized for use only by Man Lam (karmenmlam@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

Yale University Investments Office: August 2006 807-073

3

David Swensen and the Investments Office in 2006 In 1985, Swensen was hired to head the Investments Office. William Brainard, Yale’s provost at

the time, and James Tobin persuaded their former student—Swensen had earned his Ph.D. in economics at Yale in 1980—to leave his post at Lehman Brothers. The position offered not only the opportunity to help Yale but the possibility of some teaching at Yale College as well.

In the succeeding two decades, Swensen built the capabilities of the Yale Investments Office. Most importantly, he recruited and developed a very high-quality internal staff. Dean Takahashi, whom Swensen had known as a Yale student, was recruited into the Investments Office and had become Swensen’s primary lieutenant. The two worked extremely closely together. In fact, in the preface to Swensen’s book, Pioneering Portfolio Management, he described the contents as his and his colleague’s “joint intellectual property.” While a number of the Investments Office staff had gone off to lead other investment offices, such as those of the Carnegie Corporation, MIT, and Princeton, a core group of staff had worked in the office for a decade or longer, often having been hired as recent graduates of Yale College. There were a total of 24 employees (20 professionals) in the office in August 2006. Swensen encouraged his staff to be active members of the larger Yale community, and he had chosen his office’s near-campus location to signal that the Investments Office was an integral part of the University and its financial management function.

Swensen defined the role of the Investments Office broadly. Reporting to the president and to an Investment Committee (described below), the Investments Office had overall responsibility for endowment matters. While most of its day-to-day activities involved evaluating, selecting, monitoring, and overseeing external investment advisers, it also played a critical role in the entire policymaking process. For example, it was responsible for recommendations on both the investment policy and the spending policy for the endowment—that is, in broad terms, how the money should be invested and how much of it could be spent in any given year.

The Investment Committee, to which the Investments Office reported, was composed of influential and knowledgeable Yale alumni, a number of whom were quite active in different segments of the asset management business. The committee as a whole functioned as an active, involved board, meeting quarterly and providing advice to, counsel to, and ultimately approval of the various investment managers. In addition, Swensen often consulted with individual members of the Investment Committee on issues within their areas of specific expertise. This helped guide the thinking and recommendations of the Investments Office on various key issues, and it fostered an atmosphere of advice and support within which the Investments Office could take quite different and sometimes unconventional stances if it believed in them and could convince the Investment Committee of their merit.

Investment Philosophy Perhaps the most fundamental difference between Yale and other universities was its investment

philosophy. Swensen was fond of quoting John Maynard Keynes’s maxim that “worldly wisdom teaches us that it is better for reputation to fail conventionally than to succeed unconventionally.”4 Nonetheless, Swensen was willing to take “the risk of being different” when it seemed appropriate and potentially rewarding. By not following the crowd, Yale could develop its investment philosophy from first principles, which are summarized below.

4 John M. Keynes, The General Theory of Employment Interest and Money (New York: Harcourt Brace, 1936), Chapter 12.

This document is authorized for use only by Man Lam (karmenmlam@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

807-073 Yale University Investments Office: August 2006

4

First, Swensen strongly believed in equities, whether publicly traded or private. He pointed out that equities are a claim on a real stream of income, as opposed to a contractual sequence of nominal cash flows (such as bonds). Since the bulk of a university’s outlays are devoted to salaries, inflation can place tremendous pressure on its finances. Not only do bonds have low expected returns relative to more equity-like assets, but they often perform poorly during periods of rising or highly uncertain inflation. To demonstrate convincingly why he believed in the long-run advantages of equity investing, Swensen would often refer to the actual cumulative long-run returns over past decades. An original $1 investment in December 1925 in large-company U.S. stocks would be worth $2,657 by the end of 2005, and one in small-company stocks $13,706; a comparable investment in U.S. Treasury bonds would be worth $71, and one in Treasury bills $18.5 More than 95% of the endowment’s assets were expected to produce equity-like returns.

A second principle was to hold a diversified portfolio. In general, Yale believed that risk could be more effectively reduced by limiting aggregate exposure to any single asset class, rather than by attempting to time markets. While Swensen and his staff usually had their own informed views of the economy and markets, they believed that most of the time those views were already reflected in market prices. They thus tended to avoid trying to time short-run market fluctuations and would overweight or underweight an asset class only if a persuasive case could be made that market prices were measurably misvalued for understandable reasons.6

A third principle was to seek opportunities in less efficient markets. Swensen noted that over the past decade, the difference in performance between U.S. fixed-income managers in the 25th and 75th percentiles (of their performance universe) was only half a percent per year, and the difference in performance between U.S. large-capitalization stock portfolio managers in the 25th and 75th percentiles was 2% per annum. This gap widened as one moved into less liquid assets: 7.1% for hedge funds, 13.7% for buyout funds, and 43.2% for venture capital.7 This suggested that there could be far greater incremental returns to selecting superior managers in nonpublic markets characterized by incomplete information and illiquidity, and that is exactly what Yale endeavored to do. As a result, only one-third of Yale’s investment was in liquid investments such as public stocks and bonds.

Fourth, Swensen believed strongly in utilizing outside managers for all but the most routine or indexed of investments. He thought these external investment advisers should be given considerable autonomy to implement their strategies as they saw fit, with relatively little interference from Yale. These managers were chosen very carefully, however, after a lengthy and probing analysis of their abilities, their comparative advantages, their performance records, and their reputations. The Investments Office staff was responsible for developing close and mutually beneficial relationships with each of these external managers. The staff prided themselves on knowing their managers very well, on listening carefully to their ongoing advice, and on helping to guide them, if and when appropriate, on various policy matters. From time to time, the Investments Office effectively “put a team in business” by becoming a new manager’s first client. And it was not uncommon for managers to consider Yale as one of the most important of their clients.

5 R.G. Ibbotson Associates, Stocks, Bonds, Bills and Inflation (Chicago: R.G. Ibbotson Associates, 2006). Expressed in inflation- adjusted (1925) dollars, the investment in small-capitalization stocks would be worth $1377, that in Treasury bills $1.79. 6 Yale actively rebalanced its portfolio to maintain its target asset allocations, however, and this led to frequent short-term adjustments in its holdings. For instance, as equity values rose in the summer of 1987, Yale sold stocks in order to return to its target allocation level. After the stock market crash later that year, the endowment repurchased many of the same securities as it sought to raise its asset allocation back to the target level. 7 Yale University Investments Office, The Yale Endowment 2005, New Haven, 2006, pp. 36–37.

This document is authorized for use only by Man Lam (karmenmlam@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

Yale University Investments Office: August 2006 807-073

5

Finally, the Yale philosophy focused critically on the explicit and implicit incentives facing outside managers. In Swensen’s view, most of the asset management business had poorly aligned incentives built into typical client-manager relationships. For instance, managers typically prospered if their assets under management grew very large, not necessarily if they just performed well for their clients. The Investments Office tried to structure innovative relationships and fee structures with various external managers so as to better align the managers’ interests with those of Yale, insofar as that was possible.

Recent Asset Allocation and Performance Results Yale’s Investment Committee annually reviewed its endowment portfolio to decide on target

allocations to the various asset classes. The actual allocations in recent years are shown in Exhibit 1, which illustrates the recent upward trend in the allocation to the private equity, real-assets, and absolute-return classes, as well as the current (2006) target allocations. The comparable asset allocations for several groups of university endowments are shown in Exhibits 2 and 3. Private equity allocations for large institutions (including both pension funds and endowments) are shown in Exhibit 4.

As a part of the planning process, the Investments Office had completed a mean-variance analysis of the expected returns and risks from its current allocation and compared them with those of past Yale allocations and the current mean allocation of other universities. These computations, which relied on specific assumptions about the expected returns, volatilities, and correlations among asset classes, posed several issues. First, because these relationships could change dramatically over time, the Investments Office did not just rely mechanically on historical data but instead modified the historical numbers based on its own experience. Second, the Investments Office imposed limits on the amount that could be invested in each asset class. If it did not, the optimization program would instruct Yale to hold no domestic equities (or even to short-sell this asset class) and to instead invest in the more illiquid alternatives. This result followed naturally from the assumptions of the model: for instance, private equity was projected to have nearly twice the real return of U.S. equities (11.4% vs. 6.0%), albeit with a higher standard deviation (29.1% vs. 20.0%). (Over the past decade, actual returns had been considerably higher for both asset classes, and standard deviations—measured quarterly—lower.) The imposition of these constraints reflected the need of the university to diversify its holdings as well as the substantially greater imprecision with which Yale could assess the risk and return of its alternative assets. The results of this comparative mean-variance analysis are shown in Exhibit 5.

In addition, the Investments Office examined the long-run implications of its allocation for the downside risk to the endowment. In keeping with a quantitative format for analyzing long-run downside risk that had been used on prior occasions, the office examined the probability that the available endowment spending would fall by more than 10% (adjusted for inflation) over a five-year period; the office also examined the probability that the inflation-adjusted value of the endowment would fall by more than half over the next 50 years. To undertake this analysis, the Investments Office employed a probabilistic Monte Carlo analysis, which simulated and compiled thousands of possible random outcomes drawn from an assumed distribution of returns and correlations used in the simpler mean-variance analysis. This downside-risk analysis suggested that the probability of a 10% spending fall within any five-year period was 20% and that of a 50% purchasing power fall over a 50-year horizon was 15%.

Yale’s allocation philosophy and distinctive approach to investing had paid off handsomely. In fiscal year 2006 (which ended on June 30 of that year), the fund had returned 22.9%. The return

This document is authorized for use only by Man Lam (karmenmlam@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

807-073 Yale University Investments Office: August 2006

6

compared favorably with the 10.2% rise in the Wilshire 5000 Stock Index and a 1.7% fall in the fixed- income benchmark. This performance was above the average of Yale’s large peers (Columbia, Harvard, MIT, Princeton, and Stanford) at 19.3%, as well as all universities (a mean of 13.5%). Even more impressive had been the fund’s long-run performance since Swensen and Takahashi arrived at Yale. Over the 20 years ending in June 2006, Yale’s annualized return was 15.4%, exceeding the return of all colleges and universities. This result was more than 3.8% per annum better than Yale’s peers (other nontaxable endowments with over $1 billion in assets) and 5.7% per annum better than the average of all such endowments.8 (The endowment’s performance during recent years is compared with that of other universities in Exhibit 6; a more detailed breakdown of Yale’s returns by asset class is reported in Exhibit 7.) Yale’s record placed it in the top 1% in SEI’s rankings of large institutional investors.9 Not only had the average return been high, but the endowment produced consistent results: the university had not had a negative return since 1988.

The primary reason for Yale’s superior long-term performance record had been the excess returns generated by the portfolio’s active managers. Manager selection accounted for more than half of the superior performance by Yale relative to the average endowment over the last five years. As expected, the endowment’s excess returns had been greatest in the least efficient markets. Over the 10 years ending in June 2006, the annualized differences between Yale's asset-class returns and related benchmarks were 0.2% in the most efficiently priced asset class, bonds, and 12.2% in what is probably the least efficient market, private equity.

The Investments Office and the Investment Committee had been pleased with these results. As their experience with the distinctive approach grew, and they became more confident in their ability to produce sustained above-average results, they adjusted their spending policy upward. In 1992, in response to an Investments Office recommendation, the Yale Corporation adjusted the university’s long-term target spending rate upward from 4½% to 4¾% of endowment assets; in 1995, it adjusted the rate upward again, to 5%, and in 2004 to 5¼%.10 The university was thus benefiting from the strength of its investment program in two ways, both from a larger endowment and from the justified increase in the target spending rate. The substantial endowment also played a role in Yale receiving the highest rating to finance capital projects (AAA/Aaa) from the two leading bond-rating agencies and in the university’s ability to borrow money at extremely favorable interest rates.

The Management of Marketable Securities The investment philosophy outlined above guided Yale’s management decisions in all of its asset

classes. For example, Swensen and Takahashi approached bonds with skepticism. They viewed the endowment’s current target allocation of 4% in bonds primarily as a disaster reserve, guarding against a severe drop in asset values and/or deflation (such as in the Great Depression). Yale held U.S. government issues (almost exclusively): Swensen was skeptical about whether returns from U.S. corporate bonds adequately compensated investors for the added default risk and the callability of corporate issues. He was quite skeptical of foreign fixed-income securities as well. 8 All university benchmark data in this paragraph is from the National Association of College and University Business Officers, except for the 2006 data (the comparable school performance is Yale’s calculation; the average of all schools is from Cambridge Associates). Had the Yale endowment generated investment performance over the previous 20 years at the equal- weighted average of all university endowments, the endowment in June 2006 would have been $12.5 billion smaller. 9 Corporate defined benefit plans with an excess of $100 million in assets. 10 The amount of the endowment spent each year was based on a simple formula, namely, the spending rate (5¼%) times the current value of the endowment, with a 20% weight, and the value of last year’s spending, with an 80% weight; both elements are increased by inflation. In 2006, Yale spent $613 million from the endowment, representing one-third of its net revenues.

This document is authorized for use only by Man Lam (karmenmlam@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

Hailey Lam
Highlight
Yale University Investments Office: August 2006 807-073

7

Unlike most of the rest of its portfolio, the Investments Office managed its bond portfolio internally. Swensen believed that the government bond market was so efficient, and the spread between the performance of government bond fund managers so small, that it did not make sense to hire an outside manager. (In many cases, managers sought to boost returns by buying riskier bonds, thereby defeating the purpose of holding bonds in the first place.) The portfolio was managed with no attempt to add value through trading on interest rate movements. The endowment staff attempted to generate incremental returns only through modest security selection bets—for example, by using private placements issued by the Private Export Funding Corporation (PEFCO), which were backed by the full faith and credit of the United States.

Yale also owned a substantial number of U.S. common stocks, though the current target allocation, 12% of assets, was surprisingly small relative to almost all other large institutional investors. Although Yale had been an early adopter of indexing in the mid-1990s, as the Investments Office staff became increasingly confident in their ability to find superior managers it eliminated the passive portfolio in favor of a small number of active equity managers. These managers shared several characteristics. First, the majority of Yale’s active equity managers tended to emphasize disciplined approaches to investing that could be clearly articulated and differentiated from others. Swensen and Takahashi were convinced that disciplined fundamental-based approaches, when intelligently applied, could generate reliable and superior long-run performance. There were, in addition, several small stock-picking firms among Yale’s managers, firms that specialized in a very particular industry or type of investing—for example, a technology-specialist fund, one specializing in North American oil-and-gas firms, and another that held only biotechnology stocks. Not surprisingly, none of Yale’s managers tended to emphasize market timing, nor did they emphasize fuzzy or intuitive investment approaches that were difficult to articulate. The university’s managers tended to be smaller independent organizations that were owned by their investment professionals. Other things being equal, Yale preferred managers willing to coinvest and to be compensated commensurate with their investment performance. Swensen and Takahashi worried that money managers working at many organizations tended to emphasize growth in assets at the expense of performance and that ownership by a large institution reduced organizational stability and dampened incentive to perform.