Article Highlights:

• The goal of forming expectations is to make good-enough investment decisions to increase the chances of meeting life goals.
• Historically, the rolling period returns for the S&P 500 index have averaged slightly more than 10% per year, but the typical investor’s experience is different because of the high variability of outcomes.
• Beyond considering return, volatility and correlations, investors should set expectations for how much control they are willing to forsake to allow their assets to grow unimpeded.

“To achieve satisfactory investment results is easier than most realize; to achieve superior results is harder than it looks.”
—Benjamin Graham

“Everything should be as simple as it can be, but not simpler.”
—Albert Einstein

The trajectory of a happy life is shaped by expectations.

When the future meets or exceeds our expectations, we tend to be happy; when it doesn’t, we’re not.

It’s more complicated than that, but not by much. Expectations management sits at the center of a life well-lived.

So, too, with our portfolios. Infinite data provide an embarrassment of riches, but the conversation between buyers and sellers of investments is largely impoverished because we rarely put in the time and energy to simplify the key dimensions along which we set the expectations for investment outcomes. This article attempts to do just that.

There are just four foundational elements of any potential investment for which we must form expectations. These are: Returns, volatility, correlation and liquidity. Each variable invites significant complexity. Like engineers constructing buildings or bridges, experts in finance can approach investment decisions with a comparable level of specificity and precision. Collapsed bridges are not acceptable; nor are broken portfolios.

The analogy is easy, but overwrought. In fact, building portfolios to succeed in future unknown market conditions is not an engineering problem per se, but one that lends itself to a “good enough” approach. We want our portfolios to achieve what Benjamin Graham calls “satisfactory” results. Normal investors—individuals (and their advisers) whose primary concern is the ability to underwrite their life goals—are solving an expectations problem as distinct from an optimization problem [see Meir Statman, “Finance for Normal People: How Investors and Markets Behave” (Oxford University Press, 2017).]

The goal here is to make good-enough investment decisions to increase the chances of meeting subjective life goals. In this framework, there is no singular “right” answer to build the “best” portfolio.

Step one is translating the four foundational elements into more intuitive yardsticks:

• Returns → Growth
  
• Volatility → Pain
  
• Correlation → Fit
  
• Liquidity → Flexibility
  

Doing so is not just rhetorical flair. To the contrary, the move away from technical jargon to “softer” language not only eliminates intimidating terminology that stifles productive client/adviser conversation, but it also allows us to address the behavioral imperatives of better decision-making.

The main thesis here is that we can create a better investment experience by setting reasonable expectations for ranges of potential outcomes, accepting the unavoidable randomness and uncertainty that comes with making decisions about future unknowns.

Growth

We buy financial assets to meet our daily obligations and fund our dreams. What the financial literature calls “return on capital,” we’ll just call growth: We want something smaller to grow into something bigger. In addition to income, growth is the primary “product” we buy from investment companies.

What are reasonable expectations for the growth of a stock portfolio? (For the purposes of this article, we’ll focus only on stocks, but we can go through the same exercise for fixed income or other asset classes.) The answer must rely on ranges of outcomes, not pinpoint estimates. To understand why, take a look at Figure 1. It presents the average return for the S&P 500 index over rolling periods, ranging from one-year to 10-year windows (based on monthly data). As shown, over every rolling period, U.S. large-cap stocks have averaged slightly more than 10% per year.

Figure 1. Average Annual Returns by Rolling-Year Periods

Source: Copyright 2018 Ned Davis Research Inc. Further distribution prohibited without prior permission. All rights reserved. See NDR disclaimer at www.ndr.com/copyright.html. For data vendor disclaimers refer to www.ndr.com/vendorinfo.

While remarkably consistent, the data points are also highly misleading as they don’t represent the typical investor’s stock market experience. Pinpoint averages don’t inform expectations well when they are calculated from a wide range of outcomes.

When we recut the same data to look at these ranges, we end up with a very different picture, from which we set very different expectations.

Over shorter periods, the variability of outcomes has been massive, as shown in Figure 2. Across more than 1,000 rolling one-year periods from 1928 to mid-2018, the U.S. stock market climbed as much as 166% in one year and fell by 67% in another. Those large numbers are outliers, but the bulk of the range in the shorter windows is still much wider than during the longer ones. The middle two-thirds range of outcomes is also illustrated to give a sense of what has happened “most of the time.”

Figure 2. Variance in Annualized Returns by Rolling-Year Periods

Source: Copyright 2018 Ned Davis Research Inc. Further distribution prohibited without prior permission. All rights reserved. See NDR disclaimer at www.ndr.com/copyright.html. For data vendor disclaimers refer to www.ndr.com/vendorinfo.

The longer the time frame, the narrower the range of outcomes. By the rolling six- or seven-year window, the range stabilizes, thus making it easier to set long-term expectations. That said, easier does not mean easy: There is still a nearly 4,000 basis-point gap (40%) between best and worst results over those long windows. Even eliminating outliers and looking at the middle two-thirds of outcomes, there is still a large window: On a rolling 10-year basis, the outcomes vary from 4.9% to 16.3%. While both are positive numbers, the compounding effects vary markedly.

The observations so far are about the returns of the stock market. We can also set growth expectations by looking at the range of individual stock outcomes. The stock market is, well, a market of stocks, and many investors will own individual stocks rather than a broad index fund or an actively managed basket of them (e.g., mutual funds).

Consider this odd investing conundrum:

• The stock market delivers attractive gains over the long run, well in excess of cash.
  
• Most stocks don’t outperform cash over the long run.
  

In other words, most individual stocks perform just so-so to poorly, but stocks in general perform well. According to research by Hendrik Bessembinder [“Do Stocks Outperform Treasury Bills?” (Journal of Financial Economics, forthcoming)], dating back to 1926, 58% of common stocks have a lifetime return of less than a one-month Treasury bill, which is a proxy for cash. We would certainly expect that, over the long run, most stocks should outpace cash. That hasn’t been the case.

How could this be? The answer is that many companies don’t do all that well. The companies that we are most familiar with have already succeeded (a great example of “sample bias”). We rarely see the smaller companies that never climb the ladder. In fact, Bessembinder found that over the past 90 years, only 4% of companies explain the net gain for the entire U.S. stock market. The top 86 stocks (out of 26,000) accounted for 50% of the $32 trillion in wealth creation during this span. This phenomenon speaks to the critical importance of owning a diversified portfolio.

In short, equities are probably the best means for normal investors to grow their wealth over decades, but unrealistic expectations, especially over the shorter run (less than five years), will lead to disappointment and, perhaps, motivate poor decision-making.

Pain

It’s one thing to set growth expectations for one’s portfolio. It’s another to be able to hold on to that portfolio through volatile markets. For the purposes of setting expectations, volatility is the emotional cost of achieving the growth we seek. Long-term growth charts of stock and bond markets show remarkable appreciation over time. But such charts don’t give any sense of how difficult it is to hold on to our investments during the many rough patches that occur along the way. The ride matters. To properly set expectations, we must ask: How bad might it get? And am I willing to endure that pain?

One advantage here is that setting expectations is easier for pain than for growth. That’s because some asset classes are consistently more volatile than others. Stocks are more volatile than bonds. Smaller-cap stocks are more volatile than larger-cap stocks. High-yield bonds are more volatile than investment-grade bonds. Generally, the road from lending (owning bonds) to profit participation (owning stocks) grows bumpier.

Even though volatility is more predictable than returns, that jumpiness (even when known) encourages bad decisions along the cycle of greed (buying high) and fear (selling low). Volatility is the “price of admission” for access to potential growth.

For practical expectations-setting, normal investors should focus on drawdown, not volatility. Numerically, volatility is a concept with limited intuitive meaning; the stock market historically has a “vol” of around 17. Drawdown, meanwhile, is the depth of the real-world decline in any particular market, asset class or other investment. Real expectations are set by understanding the range of historical drawdowns. This is the answer to the question of “how bad can this get?”

Over the past 30 years, the averages of the top-10 drawdowns for four major U.S. asset classes were: 20.8% for large-cap stocks (S&P 500), 28.3% for small-cap stocks (Russell 2000), 12.5% for lower-quality bonds (Barclays U.S. High Yield) and 4.5% for higher-quality bonds (Barclays U.S. Aggregate). By definition, each asset class has featured drawdowns considerably worse, as shown in Table 1.

Table 1. The Biggest Drawdowns

When not properly prepared for, the jumpiness of one’s investments can lead to poorly timed selling. Consider how many investors responded to the calamity of 2008: Selling near the bottom of the market, locking in massive losses and then not reinvesting for years. Net flows into equity mutual funds didn’t turn positive until 2013.

The real-world experience of stocks has been more tumultuous than it is for bonds, as evidenced by the averages, but the range of bad outcomes shows that supposedly more conservative investments are not immune from steep losses. Note than in an expectations game, investors may be more comfortable with stock market losses than bond market losses, for bonds are supposed to serve as the ballast to a portfolio. Thus, properly managing relative expectations is critical.

Fit

As consumers, we are creatures of habit. We like what we like and tend to go back to the same vendors for the same products over and again. Cars, pasta sauce, clothes—you name it. So too with investing, where we’re drawn to a certain style or orientation of investing. For example, we see ourselves as “aggressive” or “conservative.” We like particular sectors or themes that feel right to us. That could be the industry we work in or country we live in.

Buying things we aren’t familiar with is not comfortable. Yet with investing, we’re supposed to do just that. The principle of diversification relies on owning different investments that don’t resemble each other. In fact, the less they resemble each other, the better. The more uncorrelated our portfolio investments, the more likely the portfolio should prove to be a reliable, all-weather vehicle.

For any investor—including the most sophisticated—correlation is a difficult concept. It’s a mathematically complex metric that measures the “covariance” in the prices of multiple investments. A correlation of 1.00 implies that two variables move perfectly in the same direction, whereas a correlation of negative 1.00 implies that two variables move perfectly in the opposite direction.

A smartly assembled portfolio of lower-correlated assets may provide a leg up to achieve better returns more smoothly. More growth, less pain. So, it’s worth understanding how the next investment fits into a larger plan. How does it improve upon what I already have? Is it worth it? Am I taking a distinct risk or merely doubling up on what I already own?

While mathematically precise, correlation is often unclear, even unknowable. First, the chosen time frame matters. We can measure correlation across days, weeks, months or years. It’s a somewhat arbitrary decision, but whether something is “uncorrelated” is partly a function of the time frame you choose.

Second, correlations vary over time. They are unstable. The correlations between and within categories of stocks, bonds, real estate and so forth change depending on circumstances. And here’s the rub: Correlations tend to spike higher when markets get jumpy, which is exactly what you don’t want. Correlation is a fickle friend, usually showing up for the party but rarely the funeral.

Table 2 provides a static snapshot of what correlations look like among four major asset classes: big-company stocks, small-company stocks, lower-quality bonds and high-quality bonds.

Table 2. Average Rolling Three-Year Correlations

The higher expected returns associated with factors allow investors to reduce the level of portfolio volatility they incur relative to more traditional allocations. An example is shown in the table below. The first portfolio has a 60% allocation to the S&P 500 index and a 40% allocation to five-year Treasury notes. The second portfolio has a 25% allocation to value stocks [split evenly between the Fama/French Small Value Index (ex-utilities) and the Dimensional International Small-Cap Value Index] and a 75% allocation to five-year Treasury notes. Though the 60/40 portfolio has realized a higher return, it has done so with much greater risk, as the standard deviation and worst-year return figures show. Investors seeking higher levels of expected return can increase the equity allocation in the second portfolio (from the 25% allocation) while still maintaining a comparatively greater exposure to bonds. 60% Large-Cap/ 40% Bonds 25% Value/ 75% Bonds Annualized Returns (%) 10.3 9.7 Standard Deviation (%) 10.3 7.2 Years With Returns Above/Below 15% 11/1 9/0 Years With Returns Above/Below 20% 7/0 2/0 Years With Returns Above/Below 25% 2/0 2/0 Worst-Year Return (%) -17.0 -1.4 Best-Year Return (%) 29.3 28.0 Number of Down Years 5 3 Source: “Reducing the Risk of Black Swans,” by Larry Swedroe and Kevin Grogan (BAM Alliance Press, 2018). Data © 2017 Morningstar Inc.

As with growth expectations, however, a snapshot can conceal more than it reveals.

Figure 3 shows how these ranges shift through time. You can see right away that correlations are not stable. For example, the rolling one-year correlation of the S&P 500 and the Bloomberg Barclays U.S. Aggregate Bond Index has varied significantly over the decades, even though the conventional wisdom remains that the core building blocks for a normal portfolio are stocks and bonds, precisely because they have low, even negative, correlations. That’s not always the case. In practice, it’s a moving target.

Figure 3. Rolling Correlations Over Time

Source: Copyright 2018 Ned Davis Research Inc. Further distribution prohibited without prior permission. All rights reserved. See NDR disclaimer at www.ndr.com/copyright.html. For data vendor disclaimers refer to www.ndr.com/vendorinfo.

Fit ebbs and flows. That doesn’t mean we should avoid setting expectations for it. Doing so abrogates responsibility for building a smart portfolio. However, as with the other dimensions, we need to play the tendencies over time—with clear expectations, in this particular case—for when historical correlation ranges might break down.

Flexibility

The fourth element of setting investing expectations is liquidity. Narrowly understood, this is the ease or difficulty by which you can buy or sell something. If I go to Amazon.com, I can nearly instantaneously download a book and start reading it. If I wanted to sell my house today, I couldn’t. The market for literature is liquid; the market for houses is not.

The prominence of one-click online brokerages gives the impression that investing is always cheap and seamless. Not so. Depending on the type of investment, it can range from effortless to difficult. For technicians, this is the world of bid/ask spreads and market makers. Securities become more liquid or less depending on a wide array of inputs.

Normal investors should translate liquidity into flexibility. What we really want to know, generally, is if we will be able to change our minds if we want to. If I want to pivot, can I? Will I be stuck? If so, is that a bad thing?

Being able to change our conditions lies at the center of our survival instincts. Both freedom from constraint and the freedom to chart our own course are powerful motivators. Few crave limits.

One way of correctly addressing the merits of flexibility is to ask whether you are being compensated for relinquishing it. When you sacrifice something you value, what are you getting back in return? In traditional terms of liquidity, this question is one asked every day by sophisticated investors who deal in private equity, real estate, energy partnerships and the like. Will providing long-term capital produce a return far more than what could be earned in fully liquid markets?

There’s also an important question about the behavioral compensation for inflexibility. We know, for example, when comparing outcomes for fully discretionary (e.g., a traditional brokerage account) versus retirement accounts of the exact same fund, those with less flexibility have better investor outcomes because they have been less able to quickly sell and thus better able to ride out volatile markets. For example, as Richard Thaler and Shlomo Benartzi discuss in their paper “Save More Tomorrow: Using Behavioral Economics to Increase Employee Savings” (Journal of Political Economy, 2004), investors who avail themselves of auto-investment strategies in their workplace retirement programs tend to save more and do better over time. The routine is particularly useful during bear markets, when you’re automatically investing as the market becomes cheaper.

The issue here is how much one is willing to forsake control to allow their assets to grow unimpeded. There are distinct “decision protocols” we can employ to get us toward good long-term outcomes. One is a Ulysses scenario: Tie yourself to the mast, as you know ahead of time you can’t resist the call of the sirens. In the markets, for example, you pre-commit to a strategy to stay the course (e.g., auto-invest and auto-escalate programs). You eliminate discretion. At the other end of the spectrum is retaining full flexibility and accepting that you’ll need to be patient and disciplined. The upside to this protocol is that you can change and adapt as you see fit.

In sum, while it’s hard to make an a priori judgement call that either “stay the course” or “be nimble” is better advice, investors put themselves in good stead by appreciating the trade-offs.

Conclusion: Framing Simplified Questions

The framework presented here sets simpler investment expectations. It takes the incalculable complexity of markets and boils expectations down to four—and only four—dimensions. Despite its parsimony, this will prove to be an unsatisfying exercise because (1) we’re working with ranges of possible outcomes and (2) the outcomes are inescapably uncertain. The best we can do is to play the odds and accept that the most honest answer to what our portfolios will deliver is “I don’t know.”

Such is the nature of investment decision-making—for anyone. We are ultimately trying to better manage expectations in order to motivate better behavior, for behavior is the key driver of satisfactory outcomes.

To conclude, here is a cascade of questions that investors, advisers and consultants can pose to better manage not only quantitative expectations but also the emotional discomfort associated with uncertainty:

1. What are reasonable growth expectations for my money?

• Broader: Be prepared to ask about the ranges of historical outcomes over different time frames.
  
• Deeper: Understand that inflation-adjusted, or “real,” outcomes better capture your future purchasing power.
  

2. Will the emotional discomfort of an investment’s jumpiness prevent me from enjoying its long-term growth?

• Broader: Be prepared to ask about the history of drawdowns.
  
• Deeper: Understand how the volatility of an asset class has changed over time.
  

3. Do I own a truly diversified mix of assets, including those that are lagging or appear to be “not working”?

• Broader: Be prepared to ask about how correlations might increase during market crises.
  
• Deeper: Understand how investment style tailwinds and headwinds impact the overall portfolio.
  

4. How difficult is it for me to change my mind about my portfolio?

• Broader: Be prepared to appreciate that discretion is a double-edged sword.
  
• Deeper: Work through anticipated future decisions and develop a game plan for how you would react under different circumstances.