Endowments and Foundations: Should a Windfall Change Your Investment Policy?

Brian D. Wodar

Brian Wodar and Ashley Velategui

Board members at Uptown Community Foundation (UCF) faced an enviable dilemma: what to do when an unexpected bequest boosted UCF’s capital from $10 million to $12.5 million. One board member welcomed the chance to reduce portfolio volatility by cutting the stock allocation. Another had the opposite view: with increased financial strength, why not take on greater equity exposure?

While UCF is fictional, its quandary is not unusual. Here’s how we would give a board faced with this situation a rational basis for reconsidering its asset allocation. Using our Wealth Forecasting System, we would analyze how various alternatives are likely to play out over time.

Before the bequest, UCF was sustainably spending 4.5% a year and raising $200,000 a year; it had a moderate asset allocation of 60% global equities and 40% bonds. The foundation was on track to grow its portfolio’s value over time, in median markets. The bequest made the picture even rosier.

If the foundation shifted from a 60/40 stock/bond mix to a more preservation-oriented 40/60 allocation, the nominal portfolio value would still rise, but more slowly (Display 1). In fact, it would rise so slowly that over time it would lag the level UCF would have reached without a bequest.

If the board took the opposite tack and stepped up the allocation to stocks to 80%, we project the portfolio would grow to nearly $45 million in 30 years. But this growth would come at the price of increased market risk. The larger the allocation to equities, the greater the risk of a 20% loss from a market peak to a subsequent market trough. Over the next 30 years, the odds of such a large loss would rise from just 14% with a 40/60 mix to 44% with a 60/40 mix and 75% with an 80/20 mix (Display 2). Most boards we’ve met would find the 80/20 mix too risky.

We also show boards what might happen under very unfavorable conditions—high inflation and dismal market returns. In the bottom 10% of our Monte Carlo trials, the portfolio value would decline regardless of which asset allocation they chose. Over time, truly disappointing market returns would be equally disastrous to all the investment policies studied here, including the most conservative.

Thus, both reducing and increasing the equity allocation will increase risk: the risk of permanent depletion of capital on the one hand, and the risk of a steep investment loss on the other. While different boards might reach different conclusions, maintaining the moderate asset allocation will often strike a good balance between these risks.

Adding hedge funds, non-US bonds, inflation-linked bonds, and real assets (such as real estate and commodities) is another way to improve the portfolio risk/reward trade-off that we believe merits consideration.

The Bernstein Wealth Forecasting SystemSM uses a Monte Carlo model to simulate 10,000 plausible paths of return for each asset class and inflation, producing a probability distribution of outcomes. It projects forward-looking market scenarios, integrated with an investor’s unique circumstances and taking the prevailing market conditions at the beginning of the analysis into account. The forecasts are based on the building blocks of asset returns, such as yield spreads, stock earnings and price multiples. These incorporate the linkages that exist among the returns of the various asset classes and factor in a reasonable degree of randomness and unpredictability.

Brian D. Wodar is National Director, Bernstein Nonprofit Advisory Services, and Ashley E. Velategui (CFA) is a Senior Investment Planning Analyst, both at Bernstein Global Wealth Management, a unit of AllianceBernstein (NYSE:AB).


  1. Thomas Rekdal

    In forecasting long-term equity returns is it possible to separate just the income returns? Would it ever make sense to do so, or are the total returns so closely related to the dividend income stream that it serves no useful purpose to separate them?

    An endowment fund, which in principle has an infinite investment horizon, I would think might view a 20% market decline with more equanimity if it did not imply a similar decline in income. Is this possible?

    • Brian Wodar

      Your brief question raises many relevant issues. Yes, our Capital Markets Engine accounts for total return by asset class as a combination of income and price change, so we can study endowments that spend only portfolio income. While income, price change and total return are inextricably related to each other, it is well worth studying as you’ve described so long as there is an endowment with this type of spending policy with these concerns. The Uniform Prudent Management of Institutional Funds Act (UPMIFA) provides some flexibility with spending policies, so many endowments are moving away from the traditional definition of “income,” but many still exist that spend just the portfolio income.

      To your question about market declines, if the value of the portfolio declined by, for example, 20% and had no impact on the cash yield of the portfolio, then I imagine many endowments that spend only portfolio income may be relatively comfortable with such an event. However, given that steep equity market declines are historically associated with either a concurrent or subsequent decline in cash yield, we suspect even those income-spending endowments would be just as alarmed by such a decline.

      Our experience suggests that many investors’ knee-jerk reaction to a steep market decline is to reduce or eliminate exposure to the asset class(es) that are out of favor. Public fund-flow data reflect this tendency, which means that the average investor takes “buy low, sell high” and throws it out the window at just the wrong time, over and over again.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>