The Role of Randomness & the Goal of Persistence
Sept 23, 2010 By: Mary Brunson
I spent the last week participating in an institutional investing conference. There, I had the opportunity to speak with finance directors and treasurers of organizations from all over the country. Some of the attendees are retirement plan administrators and others oversee common assets such as foundation or endowment assets.
Occasions such as this are important because they cast light on the challenges and concerns faced by the individuals who oversee institutional investments. They also provide me with a good starting point for where to begin the education process for those who are responsible for institutional assets such as retirement plans—the sort of plan you might be investing in right now.
As I ruminate about the conversations I had with many of the several hundred attendees, a couple of prominent themes prevail in my mind.
1. Each treasurer, administrator and finance director has similar needs: maximize investment returns at a specified level of risk (asset allocation), keep investments low-cost, risk-appropriate and improve transparency.
2. Many rely on the selection and termination of investment managers who actively manage their funds as an investment strategy for their endowments, retirement plans and foundations.
When we look at the goals for these investments, it’s imperative for decision makers to realize that active management is not the optimal investment strategy.
Let me explain.
The Tyranny of High Costs
Costs are an inherent component of any investment strategy. However, active management includes fees that passive management avoids—namely the weighty salaries paid to the active managers and the increased trading costs associated with the high turnover of positions in actively managed investments. Portfolios of passively managed index funds cost about one-third as much as actively managed funds. Knowing this, there is a steep hurdle that must be overcome by active management to simply keep up with its index fund counterpart—the one they are supposed to beat.
Of course, there is a supposed rationale behind the higher fees of active management. The idea is that active managers can use their stock selection acumen to ferret out above-benchmark returns. They will carefully read analysts’ reports, meet with corporate executives and assess whether a company is mispriced in a way that would allow it to appreciate faster than the overall market.
However, we find that markets are very efficient and quickly absorb the information accumulated by thousands, if not millions, of global traders into a price within minutes of the release of such information. These “fair prices” obtained by willing buyers and willing sellers are at the heart of Eugene Fama’s Efficient Market Hypothesis. Two consequences of fair prices are that investors should only expect to earn higher returns when they take higher risks and that future returns are equally likely to be higher or lower than an appropriate return for the risk.
The results of fair prices have been affirmed by the plethora of active managers who have failed to use their countless resources, abundant business acumen, and well-positioned corporate relationships to identify mispriced or “unfairly priced” equities. As the result, they have also failed in their attempt to consistently exploit the other side of their speculative trades and secure higher returns than a risk-appropriate benchmark. A recent study of 2,072 managers over a 32-year period showed that, when properly benchmarked, 99.4% of active managers lacked “genuine stock picking skill.” The chart below tells the story.
All the News that is News
When we understand the principles of the Efficient Market Hypothesis, we accept that current prices reflect known information and are the best estimates of the impact of forecasted information. In other words, everything we know about a company, about its sector and about the local and global economy has already been factored into the free and fair market price of a stock, and the only information that can move the price—either up or down—is the information that we do not have, or news which has yet to be revealed to the market as a whole. Knowing this, we understand that an active manager who seeks to cull the future strong performers and eliminate the future laggards is merely engaging in a gamble or speculation as to whether the current price already includes that prediction.
Speculation Blues
Listen to the song here.
Decision makers recoil at the idea of speculating with their investments. But, active managers are, in fact, speculating about the news that will move future prices up or down. In 1900, French mathematician Louis Bachelier told us the expected return on speculation is zero, and that is before costs. So, the expected return on speculation is negative after costs. In several active manager studies covering multiple asset classes, on average, 92% of active managers failed to beat the benchmark index. This is the cost of speculation.
Las Vegas emerged from the hot sands of the southwestern desert for one reason only—the inherent human desire to speculate about an ultimate outcome. This explains why as one travels down the Las Vegas strip, they will see signs that say ‘Our slots return 97%.’ What they really mean is, “If you give us a dollar, you have an average expectation of getting 97 cents back.” In Las Vegas, the expected return on speculation will turn a dollar into 97 cents. Despite this fact, we can still hear the ding, ding, ding of the machines and the intermittent squeals of good fortune that lure those who do not have a grasp on the real odds of success. Those fancy casinos are built on that pervasive cost of paying to play. And the television programs and magazines that track and comment on the financial markets are largely paid for by advertisers who profit from active management and often tout their recent luck. If investors really understood their odds of success, they would probably choose to not gamble at Wall Street’s really big casino. Nobel Prize winner William Sharpe asks the simple question, “Why pay people to gamble with your money?” Unfortunately, the answer to this question can best be answered by Gary Bethke who tells us, “Odds are you don‘t know what the odds are.”
Randomness vs. Persistence
When we start reading peer-reviewed academic research, we encounter the studies like “The Selection and Termination of Investment Managers.” It determined that managers who were hired by institutions actually underperformed those who were fired by those institutions over the subsequent 3-years. These results clearly show the random nature of manager performance with luck being the primary determinant of success—and luck is not a repeatable skill.
Manager selection results from the analysis of small data sets. The industry standard for fund manager reporting is 1-, 3-, 5-, and 10-year performance. With randomness driving the markets in short periods, no clear-cut characteristics are evident during these abbreviated timeframes. Statisticians tell us that we need at least 20 years before we have enough data to have a clear direction as to an expected outcome. The chart below sets forth the contrast between small sets of data vs. large sets. In the Index Comparison chart for small value vs. large growth, we see the 1- , 3-, 5-, 10-, 15-, and 20-year outcomes based on 82 years of monthly rolling period returns. These are very large data sets and provide ample data for identifying risk and return estimates of various asset classes. When such large data sets are analyzed, we begin to see the probabilities of one asset class doing better than another: a large-growth index has outperformed a small-value index in 41% of one-year periods (randomness). Large-growth has even outperformed small-value in 27% of 10-year periods. However, in 97% of 20-year periods, small-value has outperformed large-growth. This analysis of very long term history is important to determine the optimal asset allocation and what risk factors best reward investors. As legendary investor and mentor to Warren Buffett said, “In the short-term, the market is a voting machine, but in the long-term, it is a weighing machine.”
The Lessons
When you analyze long-term, style-pure index data, conclusions about the optimal investment strategy become crystal clear:
- An active manager’s short-term, above-benchmark returns are the result of lucky speculation, a non-repeatable skill. It is estimated that 3% of active managers will beat their benchmarks over long periods—through luck alone.
- Large data sets that incorporate monthly rolling period returns provide a clearer picture of the real source of stock-market returns. For equities, a risk-appropriate tilt toward small-value markets around the world carries a higher expected return.
- Markets are efficient and prices reflect all known information. Thus, stocks are fairly priced to reward investors for the risks they bear. The distribution of errors in the price looks similar to a bell curve and the errors are not known in advance.
- An institution can improve its expected returns by simply eliminating unnecessary fees and expenses. The high cost of paying active managers to gamble with your money and consultants to hire and fire those managers is great place to start the cost cutting.
- The best way to achieve higher risk-adjusted returns is simply to buy, hold and rebalance a risk-appropriate blend of low-cost, style-pure indexes that have long-term risk and return histories. When you do, you can invest and relax.
It is IFA’s privilege to share this information with you. Each of our investment professionals welcomes the opportunity to assist you in your quest for risk-appropriate, low-cost returns. To learn more, please call 888-643-3133 or visit ifa.com.