Читать книгу Alternative Investments - Black Keith H. - Страница 78

There is an important tension between the spending rate of the endowment, the risk of the endowment portfolio, and the goal of allowing the endowment to serve as a permanent source of capital for the university. When the endowment fund generates high returns with a low spending rate, the size of the endowment fund increases. This may lead to concerns about intergenerational equity, as the spending on current beneficiaries could likely be increased without compromising the probability of the endowment continuing into perpetuity. Conversely, a conservative asset allocation with a high spending rate may favor the current generation yet imperil the real value of the endowment in the long run.

Kochard and Rittereiser (2008) present a history of endowment spending models. From the founding of the Harvard University endowment in 1649 until the 1950s, endowments were typically managed conservatively, with a focus on earning income from a fixed income–dominated portfolio. It was clear what the spending rate should be, as the entire portfolio yield was typically paid out to support the programs of the university. When bonds matured at face value, the notional value of the corpus was maintained. This sounds relatively straightforward, as the current beneficiaries received a strong income, and the requirement of maintaining the nominal value of gifts made to the foundation was easily achieved. Unfortunately, this conservative asset allocation earned little in the way of real returns, and the focus on minimizing drawdowns led to a low total return and stagnation of the nominal value of endowments.

Between the 1950s and 1970s, this conservatism began to fade. Rather than spending the income generated on the portfolio, endowment managers came to embrace the concept of total return. Consider a fixed-income portfolio with a 5 % current yield, which allowed a 5 % spending rate. Though the yield was high, the nominal return to the portfolio, net of spending, was zero. In this scenario, the spending needs were easily met, yet the value of the endowment declined in real terms, as the goal of maintaining the real, or inflation-adjusted, value of the corpus was not achieved. In seeking to maximize the portfolio's risk-adjusted return, a total return investor (i.e., an investor who considers both income and capital appreciation as components of return) may realize that a 5 % current yield is not needed in order for the endowment to have a spending rate of 5 %. Moving from a portfolio dominated by fixed income to one with a healthy mix of equity investments may reduce the yield to 3 % while increasing total return to 7.5 %. With a total return of 7.5 %, including income and capital appreciation, the endowment can afford a spending rate of 5 % while maintaining the real value of the portfolio, as the 2.5 % return in excess of the spending rate can be used to offset the impact of inflation. In order to generate 5 % spending, the entire income of 3 % is spent, and 2 % of the portfolio is sold each year to meet the spending rate.

The investing behavior of endowment managers began to change between 1969 and 1972. The 1969 Ford Foundation publication Managing Educational Endowments suggested that endowment portfolios had previously invested with overly conservative asset allocations. Endowment managers were chastised for building underperforming portfolios that were underweighted in equities, not celebrated for serving the university by reducing the risk of loss. The publication proposed that trustees might have imperiled the university by forgoing higher returns that could have increased both the income and the corpus of the endowment. The 1972 Uniform Management of Institutional Funds Act subsequently allowed endowment managers to consider total return when setting the spending rate. The act also allowed the use of external investment managers and encouraged trustees to balance the long-term and short-term needs of the university.

It was during the bull market in public and private equities, from 1982 to 2000, that endowment managers dramatically increased their equity exposures. This produced extremely strong returns that boosted the values of endowments by large multiples, net of spending. The post-2000 drawdown in equity markets led endowments to have an even stronger focus on alternative investments, moving to increase exposure to hedge funds and other assets with lower correlations to equities while reducing exposure to publicly traded equities.

Once an asset allocation is determined, a spending rule must also be established. The earliest endowments spent income only, which tilted the portfolio toward income-producing securities. Later, endowments moved toward spending at a fixed percentage of the current value of the endowment, such as 4 %. This fixed spending rate, however, created volatility in the amount of income available to the university. In a year when the return and gifts received by the endowment generated a 20 % increase in endowment value, the income to be spent also increased by 20 %. Conversely, during a 20 % drawdown, the spending rate of the university was slashed by a significant amount. A sticky spending rate, such as $3 million per year, provides certainty of income to the university but can create concerns of intergenerational equity after a large gain or loss in the value of the endowment.

Recognizing that volatility in the income provided to university operations was unwelcome, more flexible spending rules were developed. Spending 4 % of the average value of the endowment over the trailing three to five years creates a smoothing process that dampens the impact of the volatility of portfolio returns on the income provided to the university.

David Swensen developed a spending rule for Yale University: each year, the endowment could spend at a rate equal to 80 % of the prior year's dollar spending plus 20 % of the endowment's long-term spending rate (4.5 %). This formula incorporates the prior 10 years of endowment value into the spending rate calculation, providing a stronger smoothing effect than a simple moving average rule.

Swensen (2009) expresses concern about the impact of inflation, market volatility, and high spending rates. Some 70 % of endowments use spending rates of between 4 % and 6 %, with 5 % being a popular choice. Yale has set its long-term spending rate at 4.5 %, as simulations show that, based on a five-year average of endowment values, a 5 % spending rule has a 50 % probability of losing half of the endowment's real value at some point over the course of a generation. Using a longer averaging period and a lower spending rate reduces the probability of this disastrous decline in the real value of the endowment portfolio.

Inflation has a particularly strong impact on the long-term real value of university endowments. Ideally, endowments should seek to maintain the real value of the corpus rather than the legal requirement of the notional value. When maintaining the real value of the corpus, long-term spending rates can keep up with inflation; when maintaining the nominal value of the corpus, long-term spending has an ever-declining value in real terms.

The focus on inflation risk has led many endowments to increase allocations to real assets in recent years. Between 2002 and 2014, the largest endowments increased their average allocations to real estate and natural resources from 6 % to 14 %. Interestingly, while endowments invested a greater portion in real estate than in natural resources in 2002 (4.3 % versus 1.7 %), they are currently investing the same amount in both natural resources and real estate (7.0 %). Real asset investments include inflation-linked bonds; public and private real estate investments; commodity futures programs; and both direct and private equity fund investments in mining, oil and gas, timber, farmland, and infrastructure. Ideally, the real assets portfolio would earn long-term returns similar to those of equity markets, with yields similar to those of fixed income, while experiencing low volatility and low correlation to the fixed income and publicly traded equity assets in the portfolio, as well as higher returns during times of rising inflation.

A report from Alliance Bernstein (2010) calculated the inflation betas of several asset classes. An inflation beta is analogous to a market beta except that an index of price changes is used in place of the market index, creating a measure of the sensitivity of an asset's returns to changes in inflation. Just a few assets demonstrated a positive inflation beta, where the assets act as an effective inflation hedge. The majority of assets have a risk to rising inflation; that is, they have a negative inflation beta. According to Alliance Bernstein, commodity futures offered the greatest inflation beta at 6.5, whereas farmland had a beta of 1.7. In the fixed-income sector, 10-year Treasury Inflation-Protected Securities (TIPS) had a beta of 0.8, whereas three-month Treasury bills had a beta of 0.3. Equities and long-term nominal bonds had a strong negative reaction to inflation, with the Standard & Poor's (S&P) 500 Index exhibiting an inflation beta of –2.4 and 20-year U.S. Treasury bonds suffering returns at –3.1 times the rate of inflation. Within equities, small-capitalization stocks had an even greater risk to rising inflation. Companies with lower capital expenditures and fewer physical assets also had a stronger negative response to rising inflation.