• Basic Membership Access to Journal, Shadow Stock portfolio & monthly updates for stock screens
  • A+ Investor Full access to AAII.com, Journal, and a power data suite for investment ideas
  • AAII Platinum Full access to all of AAII's model portfolios and premium services at a significant discount
  • Dividend Investing Newsletter and model portfolio that focuses on dividend growth
  • AAII Growth Investing Find secular growth stocks poised for outperformance regardless of macroeconomic conditions.
  • Retirement Investing Discover trusted strategies for every stage of retirement. Get actionable insights and resources for smart retirement planning.
  • Sentiment Investing A research-driven newsletter with 10+ signals decoding sentiment across retail and institutional behavior.
  • Stock Superstars Report Get stock ideas inspired by the market's leading professionals
  • VMQ Stocks For the value investor looking to harness academic research in an actively managed portfolio
  • Bits and Bips Newsletter delivering insights on linking traditional finance, macro trends and cryptocurrency opportunities to help investors build stronger, more diversified portfolios.
All premium products come with free basic membership.
Explore All Membership Options
Join Now
Asset Allocation Portfolio Strategies Risk

Why You Should Consider Volatility When Building a Portfolio

One observation from looking at six model portfolios is that focusing on the volatility of your portfolio gives you a more reliable measure of how it is actually behaving.

by Craig Israelsen | October 2022

Featured Tickers:

Portfolios are as varied as the people assembling them, so it’s challenging to “study” all the possibilities. Nevertheless, we can study “representative” portfolios to get a sense of how they behave over time. Toward that end, I’ve built six different portfolio models ranging from 100% stocks to 100% bonds. And to make things more interesting I’ve built a Vanguard version, a T. Rowe Price version and a “variety pack” version.

A 60% equity/40% fixed-income portfolio is highlighted in Table 1.

TABLE 1 All-Stock to All-Bond Portfolio Allocations Portfolio allocations for each of the six portfolios. Each portfolio was created using Vanguard funds, T. Rowe Price funds and funds from a variety of other fund companies.

The Vanguard version of the 60%/40% model has a 15% allocation to the Vanguard Short-Term Federal fund (VSGBX), a 25% allocation to the Vanguard Total Bond Market Index fund (VBMFX), a 30% allocation to the Vanguard 500 Index fund (VFINX), a 15% allocation to the Vanguard International Growth fund (VWIGX) and a 15% allocation to the Vanguard Explorer fund (VEXPX). The allocations to each fund are maintained by monthly rebalancing (not recommended in actual practice but was necessary in this particular analysis).

The 60%/40% T. Rowe Price model has the following allocations: 15% to the T. Rowe Price Short-Term Bond fund (PRWBX), 25% to the T. Rowe Price New Income fund (PRCIX), 30% to the T. Rowe Price Equity Income fund (PRFDX), 15% to the T. Rowe Price International Stock fund (PRITX) and 15% to the T. Rowe Price New Horizons fund (PRNHX). Finally, the variety-pack portfolio utilizes funds from Dimensional Fund Advisors (DFA), Dodge & Cox, Invesco, Templeton and Wasatch Global Investors to build the six different portfolios ranging from all-stock to all-bond.

Portfolio Returns and Volatility

Let’s now examine how these various portfolios performed over the past 32 years (January 1990 to December 2021) as shown in Table 2.

TABLE 2 Standard Deviations and Returns 1990–2021 Standard deviation (volatility) and returns for each portfolio for the period of January 1990 through December 2021. The results assume that a lump-sum investment was made at the beginning of the period and each portfolio was rebalanced monthly.

We begin with the 100% equity portfolios. If using the three Vanguard funds shown in Table 1, the 32-year annualized standard deviation of monthly returns was 15.34% and the 32-year annualized return was 10.43%. If using T. Rowe Price funds, the 32-year standard deviation was 14.71% and the 32-year return was 10.24%. The variety-pack portfolio had a standard deviation of 15.08% and a return of 10.36%. The volatility (as measured by standard deviation) and the return of each 100% equity portfolio was remarkably similar between the Vanguard, T. Rowe Price and variety-pack models. Key observation here: Asset allocation is the primary driver of portfolio behavior, not where we buy the funds from.

Consider an 80% equity/20% fixed-income portfolio. The Vanguard version had a 12.31% standard deviation and a 9.58% average annualized return over the past 32 years. The T. Rowe Price model’s standard deviation was 11.93% with a return of 9.37%. The variety pack had a standard deviation of 12.26% and a return of 9.62%. Same observation: strikingly similar levels of volatility and return over a 32-year period. In fact, we see the same pattern in all six asset allocation models—the volatility and performance among the Vanguard, T. Rowe Price and variety-pack models are nearly in lockstep.

There is another key observation in Table 2 that nearly jumps off the page: The differences in standard deviation (aka “risk” or “volatility”) among the various allocations are far greater than the differences in returns. For example, the 100% equity portfolios have a standard deviation of roughly 15% and the 100% fixed-income portfolios have a standard deviation of approximately 3%. That is a 5x difference.

Now, look at performance. The 100% equity portfolios have a return of roughly 10.3% and the 100% bond portfolios generated an annualized return of roughly 5%. That’s basically a 2x difference. In other words, the asset allocation we choose to employ has more potential impact on volatility than it does on performance. Said differently, we have more control over volatility (by our chosen asset allocation) than we do over performance.

Here’s another way to think about this: Over the past 32 years, if we moved from an all-bond to an all-stock portfolio our performance basically doubled, but our volatility increased by a factor of five. We do have to remember that between 1982 and 2021, interest rates were generally in decline. The drop in interest rates was a real boon to U.S. bond funds. Going forward, bonds may not do as well. This could cause the performance gap between an all-bond and an all-stock portfolio to increase in the future.

The Stability of Volatility

Figure 1 shows a rolling 36-month standard deviation and rolling 36-month returns of the Vanguard 60%/40% asset allocation model. The standard deviation is shown by the blue line and the returns are shown by the orange line. Which one is more volatile over rolling 36-month time frames? Rolling returns are far more volatile than standard deviation.

FIGURE 1 Volatility Versus Returns 1990–2021 The returns of the Vanguard 60%/40% portfolio have been more volatile than the standard deviation of the same portfolio over the last 32 years (1990–2021).

Rather ironic, isn’t it? Standard deviation over rolling 36-month periods tends to operate in a smaller “bandwidth” of roughly 5% to 15%. This is the case because standard deviation of return can never be negative, whereas performance can be negative (as shown by the 2000–2002 and 2006–2008 time frames). By comparison, rolling 36-month returns ranged from –8% to +19%.

What is the implication of this observation? I would propose that portfolios be built to achieve a given level of volatility. (The results in Table 2 provide a starting point for expected levels of volatility and performance for various asset allocation models.) Your expected volatility and your actual experienced volatility will be much closer than your expected return and your experienced return—particularly over short time frames such as three years or less. Even though investing is a long-term commitment (i.e., 20 years or more), most of us check our portfolios far more often than that, so volatility and performance over shorter time periods do matter.

As investors, if we focus on the volatility of our portfolio, it likely won’t shock us in any given three-year period. Portfolio volatility is fairly consistent over time. Alternatively, if we focus solely on performance, there will be three-year periods where returns and our portfolio balance will be far from our expectations—and that can be a bit shocking.

Thus, focusing on the volatility of our portfolio will give us a more reliable measure of how our portfolio is actually behaving—and how volatile it isn’t. 

Discussion

BARRY J from TX posted over 3 years ago:

If I read “Investing at Level 3” by AAII founder James Cloonan correctly, Cloonan recommends AAII members ignore SHORT-TERM volatility COMPLETELY. I don’t remember reading in Mr. Israelsen’s article reading any explicit statement that his objective is to support Cloonan's advice. However, Israelsen achieves that tacitly by showing AAII members how they can MINIMIZE the distraction of constant variation in their portfolios by REDUCING it up front when they design their portfolios. Great so far. But at what price? I have more thoughts on this to share after I finish my research.


ROBERT A from NC posted over 3 years ago:

I find many of Mr. Israelsen's articles to be informative, but this one has left me scratching my head. Mr. Israelsen never adequately answers the article's title. He states, "I would propose that portfolios be built to achieve a given level of volatility," but he never really explains WHY. WHY on earth should I care about volatility when my goal is to maximize my wealth in the long run? The 32-year average annual return tells me all I need to know. If I want to maximize my wealth over those 32 years, I'll invest 100% of my money in equities. Mr. Israelsen also states, "The 100% equity portfolios have a return of roughly 10.3% and the 100% bond portfolios generated an annualized return of roughly 5%. That’s basically a 2x difference." No, no, NO! Over 32 years, 10.3% annualized is nearly FIVE TIMES as much as 5% annualized. That's the difference between having a retirement fund of $1million versus $5million! I've been 100% invested in equities all my adult life, and I'm glad I never listened to the "volatility is risk" crowd.


GREGORY T from WI posted over 3 years ago:

Volatility is measured by standard deviation: standard deviation of what? Does volatility take into incorporate dividend payments, or is it just the standard deviation of the price? If so, is it done quarterly or on some other frequency?


RICHARD H from IL posted over 3 years ago:

This article covers an well-worn path addressed by Walter Shewhart in the 1920's called SPC or statistical process control. For example, the finding that the moving average varies less than the standard deviation with 32 observations is a variant of the central limit theorem that says that the standard deviation of the average is merely the standard deviation of the process divided by the square root of the number of samples. . . in this case, approximately 5.6, which is in the same ballpark observed by Craig Israelsen. The use of SPC in stock selection and planning is well known by quality professionals because it helps in identifying when a special cause might be influencing the price targets and standard deviations (spreads) either up or down. Unfortunately, special causes (also popularly known as "black swans events") result in resetting the average, target, and spread. In industry, we work hard to remove the special causes so they don't affect future measurements. It's difficult there even though industry has control over their processes. That isn't the same situation in the stock market. In the stock market, this approach can alert us to the presence of a "black swan" but it remains up to us to identify the root cause of the change AND then prognosticate the effect that cause will have on the price.


BARRY J from TX posted almost 2 years ago:

Dr. Israelsen, as promised, I reread this article (two years later in 2024) after re-educating myself on the mathematics involved in your analysis. Harry Markowitz’s Ph D explained why he selected the relationship between return and volatility and why he chose SD as a proxy to measure both market volatility and portfolio covariances to be USD to select the most appropriate position along “the efficient frontier” to understand random processes like capital markets. Willaim Sharpe’s Ph D on CAPM reinforced the justification of using SD as a proxy for measuring variation in markets because they are random processes (as later reported by Gene Fama) and extended Markowitz’s justification to pricing options. A key step in convincing professional investors of this math's value is understanding asset pricing. In “The Intelligent Asset Allocator” (2021) Willaim Berstein confirmed your findings and demonstrated the same principle that – you will maximize portfolio return if you accept (“tolerance”) a minimum asset covariance across asset allocations for that return. This is analogous to the process in your article to achieve “genuine diversification.” Thank you for pointing me along the path of enlightenment.


You need to log in as a registered AAII user before commenting.
Create an account

Log In

Get your free copy of our special report analyzing the tech stocks most likely to outperform the market.
Download the FREE Report Here: