Return to:Previous Page
Morningstar.com's Interactive Classroom

Course 203
Looking at Historical Risk, Part 1

Introduction

Most natural risk-takers mountain climbers, extreme athletes, motorcycle daredevils tend to talk about the high of a job well done, an adventure completed, a successful free fall, and so on. They're less likely to dwell on bones broken, damaged equipment, and the heavy insurance costs that surely dog them.

Investors tend to behave a little like these extreme athletes, at least when they're starting out: They would much rather talk about the returns their funds generated than the risks they took to achieve those returns or the losses they've incurred. Take the large-cap Cambiar Opportunity Fund for example. This fairly concentrated fund landed in the top 10% of its category in 2001-2004, but reversed course in 2007 and 2008, landing in the category's worst 5% and 25% respectively, before returning to its winning ways in 2009 and 2010. Tremendous gains are won only through tremendous risk taking, which often means many ups and downs in short-term returns. That's called volatility.

While no single risk measure can predict with 100% accuracy how volatile a fund will be in the future, studies have shown that past risk is a pretty good indicator of future risk. In other words, if a fund has been volatile in the past, it's likely to be volatile in the future.

In this lesson, we'll tackle two common yardsticks for measuring a mutual fund's risk: standard deviation and beta. Both of these measures appear on a fund's Morningstar fund report page.

Standard Deviation

Standard deviation is probably used more often than any other measure to gauge a fund's risk. Standard deviation simply quantifies how much a series of numbers, such as fund returns, varies around its mean, or average. Investors like using standard deviation because it provides a precise measure of how varied a fund's returns have been over a particular time frame both on the upside and the downside. With this information, you can judge the range of returns your fund is likely to generate in the future. Morningstar calculates standard deviations for the most recent 36 months of a fund's life. The more a fund's returns fluctuate from month to month, the greater its standard deviation.

For instance, a mutual fund that gained 1% each and every month over the past 36 months would have a standard deviation of zero, because its monthly returns didn't change from one month to the next. But here's where it gets tricky: A mutual fund that lost 1% each and every month would also have a standard deviation of zero. Why? Because, again, its returns didn't vary. Meanwhile, a fund that gained 5% one month, 25% the next, and that lost 7% the next would have a much higher standard deviation; its returns have been more varied.

Standard deviation allows a fund's performance swings to be captured into a single number. For most funds, future monthly returns will fall within one standard deviation of its average return 68% of the time and within two standard deviations 95% of the time.

Let's translate. Say a fund has a standard deviation of four and an average return of 10% per year. Most of the time (or, more precisely, 68% of the time), we can expect the fund's future returns to range between 6% and 14% or its 10% average plus or minus its standard deviation of four. Almost all of the time (95% of the time), its returns will fall between 2% and 18%, or within two standard deviations of its mean.

Using standard deviation as a measure of risk can have its drawbacks. It's possible to own a fund with a low standard deviation and still lose money. In reality, that's rare. Funds with modest standard deviations tend to lose less money over short time frames than those with high standard deviations. For example, the one-year average standard deviation among ultrashort-term bond funds, which are among the lowest-risk funds around (other than money market funds), is a mere 0.64%.

The bigger flaw with standard deviation is that it isn't intuitive. Sure, a standard deviation of seven is obviously higher than a standard deviation of five, but are those high or low figures? Because a fund's standard deviation is not a relative measure which means it's not compared with other funds or with a benchmark it is not very useful to you without some context.

So it's up to you to find an appropriate context for standard deviation. To help determine if your fund's standard deviation is high or low, we suggest you start by looking at the standard deviations of similar funds, those in the same category as the fund you're examining. In May 2011, for example, the average mid-cap growth fund carried a standard deviation of 26.4, while the typical large-value fund's standard deviation was 22.5. You can also compare a fund's standard deviation with that of a relevant index. The S&P 500, a common benchmark for large-cap funds, for example, had a standard deviation of 21.7 in May 2011.

Beta

Beta, meanwhile, is a relative risk measurement, because it depicts a fund's volatility against a benchmark. Morningstar calculates betas for stock funds using the S&P 500 Index as the benchmark. We also calculate betas using what we call a fund's best-fit index, which is the benchmark whose performance most resembles that of the fund. For bond funds, for example, we use the Barclays Capital Aggregate Bond Index and best-fit indexes.

Beta is fairly easy to interpret. The higher a fund's beta, the more volatile it has been relative to its benchmark. A beta that is greater than 1.0 means that the fund is more volatile than the benchmark index. A beta of less than 1.0 means that the fund is less volatile than the index.

In theory, if the market goes up 10%, a fund with a beta of 1.0 should go up 10%; if the market drops 10%, the fund should drop by an equal amount. A fund with a beta of 1.1 would be expected to gain 11% if the market rises by 10%, while a 10% drop in the market should result in an 11% drop by the fund. Conversely, a fund with a beta of 0.9 should return 9% when the market goes up 10%, but it should lose only 9% when the market drops 10%. The biggest drawback of beta is that it's really only useful when calculated against a relevant benchmark. If a fund is being compared with an inappropriate benchmark, its beta is meaningless.

There's another statistic that is often overlooked in this discussion of volatility: R-squared, which you can find under the Ratings & Risk tab of a fund's report on Morningstar.com. The lower the R-squared, the less reliable beta is as a measure of the fund's volatility. The closer to 100 the R-squared is, the more meaningful the beta is. Gold funds, for example, have an average R-squared of just 0.26 with the S&P 500, indicating that their betas relative to the S&P 500 are pretty useless as risk measures. Unless a fund's R-squared against the index is 75 or higher, disregard the beta.

Quiz 203
There is only one correct answer to each question.

1 What does standard deviation measure?
a. When a fund's returns have beaten its peers' over a particular time period.
b. How volatile a fund's returns have been versus a benchmark over a particular time period.
c. How likely a fund is to lose money.
2 A fund has a standard deviation of 10 and an average return of 12% per year. What does that mean?
a. Sixty-eight percent of the time, the fund's future returns will range between negative 8% and 32%.
b. Ninety-five percent of the time, the fund's future returns will range between negative 8% and 32%.
c. Ninety-five percent of the time, the fund's future returns will range between 2% and 22%.
3 To make the most of a fund's standard deviation, compare it with:
a. The fund's R-squared.
b. The fund's beta.
c. The standard deviations of other funds in its category.
4 A fund with a beta of 1.20 will do what if the market falls 10%?
a. Rise 20%.
b. Fall 20%.
c. Fall 12%.
5 You can draw the most accurate conclusions about the risks of which fund?
a. A fund with a beta of 1.10 and R-squared of 97.
b. A fund with a beta of 1.15 and an R-squared of 50.
c. A fund with a standard deviation of 20 and an R-squared of 95.
To take the quiz and win credits toward Morningstar Rewards go to
the quiz page.
Copyright 2006 Morningstar, Inc. All rights reserved.
Return to:Previous Page