Leave it to Tom Hanks. Only the most acclaimed actor of our generation could have garnered an Oscar nomination for absolute silence, interrupted only by an occasional shouting match with his co-star, a bloodied volleyball named Wilson. Castaway, released in 2000, showed Hanks’s character, a pudgy executive from FedEx, marooned on a South Pacific island as the lone survivor of an airplane crash. His only companion was a volleyball that washed ashore in a FedEx package. The flick now regularly pops up on cable and remarkably always seems to draw me in, suspending my channel surfing for a few minutes. There’s just something about the drama of surviving on a deserted island1 that requires nothing more than watching what happens next.
To most, a portfolio management process without frequent performance reviews and benchmark comparisons is about as foreign as a movie without dialogue or co-stars. But most asset management programs have very long horizons, where a month or quarter—let alone a day or a week—is but a statistically insignificant blip. Actuarially, even an octogenarian has a nearly 10-year time horizon. In recent talks with advisors, I asked how many had clients whose investment portfolios were designed to meet liabilities less than 10 years out. Anecdotally, I can tell you well under 5% of the advisors raised their hands.
Of course, some institutional investment programs also are designed to meet prospective payout streams that go on for decades. This is indeed a blessing because the capital markets are inherently noisy and unpredictable over the short term, but remarkably steady and predictable over the long term. But all of us succumb to the temptation to check the impact of recent market movements on manager performance. One large pension client specifically affirmed this tendency by telling our CIO that “we are long-term investors but short-term reviewers.”
What if we couldn’t evaluate performance for the next quarter or the next year or even the next three years? At Research Affiliates, we call this thought experiment the “Deserted Island portfolio.” Like Hanks’s character, we would be exiled to a remote, uninhabited island. But before we get banished to the beach, we could build a portfolio we would buy and hold, save for an annual rebalancing back to target allocations. In the interim, there’d be no cable news feed and no benchmark comparisons, nothing to reinforce the brilliance or stupidity of our portfolio selections.
Today this exercise reveals some unusually good opportunities for those willing to embrace maverick risk.
Deserted Island Logic
In the late 1980s, Paul B. Andreassen conducted a series of experiments demonstrating that investors who received no news about their stock market investments did better than those who saw the headlines. The theory is that investors often overreact to short-term news that has little effect on the long-term fair value of the company. Some 25 years later, the 24/7 news cycle and nearly continuous performance measurement cycle makes it all the easier for investors to overreact, generally to the detriment of portfolio returns.
We are unlikely to change this dynamic, and the reason is simple. Nearly all of us in the investment business are agents rather than the actual owners of capital. We oversee others’ nest eggs, pensions, and charitable trusts, most of which have timelines measured in decades. Acting as their clients’ agents, advisors and managers of capital with discretionary authority have a very different timeline, one that is often measured in a handful of years. If our results fall short in comparison with common benchmarks and peer groups, we risk being “de-selected.”
The stark difference in time horizons is exacerbated by the fact that all of us must report to someone, that all of us have a client. In the public arena, for example, the portfolio manager reports to the chief investment officer, who reports to the chairman of the asset management firm, who reports to the client’s chief investment officer, who reports to the investment committee, who reports to the state pension board, who reports to the governor and ultimately the taxpayers. Each rung in the ladder adds pressure to deliver relative results, condensing the timeframe for evaluating one’s success or lack thereof. Accordingly, investment professionals embrace maverick risk at their own peril.
This is hardly new stuff. Dean LeBaron, in his 1983 reflections on market inefficiency, noted that investment managers often are more motivated to please the various levels of clients (with their inevitably diminishing time horizons for evaluating success) than to engage in profitable transactions. He went on to say, “Many managers end up paying a good deal in terms of foregone [sic] opportunity for the privilege of resting comfortably in the lower part of the second quartile and avoiding ever being in the bottom quartile for even a short period.”
One way to break this way of thinking is to imagine not having to report to anyone for a full market cycle. What sort of portfolio would we construct if we weren’t subject to a continual barrage of benchmark returns and peer group comparisons?
To answer, we need to develop some forecasts. Longtime readers of this publication have probably seen us walk through our “Building Blocks” approach, a simple “yield plus growth” calculation that provides an intuitively direct and empirically robust starting point for forecasting long-term asset class returns. When we apply it to mainstream U.S. equities and bonds, we find future long-term returns (5 to 10 years) are likely to be paltry—about 2.5% annually for bonds and 5.5% annually for stocks, as shown in Table 1. A typical 60/40 stock/bond mix can therefore be expected to produce 4–5% returns, starkly below the 7–8% returns typically built into pension and defined contribution assumptions.2