The New “Ab”normal: Measuring Volatility is Risky Business

Market pundits suggest we are entering into an era of the “new normal” characterized by low returns and high volatility on risky assets.  Following an earlier article in the Financial Times (FTfm supplement) Stephen Horan, head of professional education content and private wealth management at CFA Institute  and Raghu Mandagolathur of CFA Kuwait, the national Kuwait society, discuss how the financial crisis has encouraged new thinking about what it means for markets to be abnormal and how to measure it.  The GCC stock markets measure very high in terms of volatility, even higher than emerging markets. The high volatility contributes to excessive speculation which reduces market efficiency. The issue is exacerbated by index concentration with few large cap stocks dominating the index. While decline in volatility happens over a long period of time when markets become more efficient, investors would do well to understand its source and take measures to manage it.
<!--[if !supportLineBreakNewLine]-->
<!--[endif]-->Why is it important to measure market volatility?

Volatility has implications for the future market behavior.  Research shows that abnormally large price movements (whether positive or negative) tend to be followed by relatively high returns.  Investors gain confidence as markets gradually resolve uncertainty instigated by the initial price shock. 

 Moreover, price shocks portend continued high volatility in the near term.  In other words, volatility tends to cluster through time.  So, even if an investor is unable to anticipate the onset of increased market volatility, they can expect it to continue for a while once it begins. 

How do portfolio managers typically measure market volatility?

Investors often think of market volatility in terms of the standard deviation of historical returns, which measures how variable returns have been over some time period.   Options markets provide an opportunity to get a forward-looking glimpse of volatility because the market price of calls and puts is directly related to the standard deviation of returns expected to prevail over the life of the option. However, GCC region is yet to have the benefits of a fully fledged derivatives market. Kuwait has a limited options market with only call options being allowed to be written...

Returns that are one standard deviation away from average are more typical than returns three standard deviations away from average.  Whether based on historical experience or implied by option prices, an asset’s standard deviation provides a benchmark for normal, or typical, variation relative to itself.

Are we in an abnormally volatile era?

The EURO STOXX 50 Volatility index, which measures the forward-looking standard deviation implied by prices of EURO STOXX 50 index options, has been close to its 52-week low, suggesting that market participants anticipate relative tranquility in the near term.   Based on this measure, one might say we are in an era of abnormally low volatility.  All else equal, markets respond negatively when implied volatility increases.

Are there other ways to measure market volatility?

One could think about the variability of returns among individual stocks in an index rather than the variability of broad market index relative to itself.  The more disperse the returns among a group of securities, the greater the opportunity for profitable security selection within that group.  This volatility across securities tends to spike in times of crisis, suggesting that returns from active management can be most pronounced during these times.

In a more eclectic vein, market volatility can be measured using linguistics to quantify the volume, tenor, and variability of sentiment of newswire stories.  This non-traditional, forward-looking approach can be helpful when forward-looking implied volatilities from option contracts are unavailable. 

Finally, volatility can be measured in terms of how asset classes typically behave in relation to one another, sometimes called market turbulence.

What do you mean by “turbulence”?

For example, in relatively tranquil markets, returns on corporate bonds and large company stocks are normally fairly independent of each other.  In turbulent markets, the returns of these asset classes may tend to move together. 

Even if return variability of corporate bonds or large company stocks is not particularly abnormal based on their individual behavior, the market can be turbulent based on the uncharacteristically high correlation between the two asset classes.  Mark Kritzman, CFA, and Yuanzhen Li propose a method measuring turbulence that incorporates unusual correlations in a recent Financial Analysts Journal article.

Are you simply describing the notion that correlations increase in times of crisis?

The concept has broader applicability.  In some cases, correlations decrease in turbulent markets.  For example, returns on high-yield corporate bonds and high-quality government bonds tend to be highly correlated in normal markets when returns on fixed income instruments are dominated by interest rate movements. 

In times of stress, changes in the risk premium dominate price changes, driving down returns on high-yield bonds and pushing up returns on high-quality government bonds in a flight to safety.  The main point is that turbulence is defined as much by abnormal correlation of returns as it is by abnormal volatility.

How might portfolio managers integrate different notion of volatility into practice?

First, portfolio managers can use the variability of returns across different securities to judge the profit potential of active management in a particular asset class and adjust their active or passive strategy weights accordingly. 

Portfolio managers might also spend extra time conducting scenario analyses based on experiences in market crises and focusing on variables that have been historically unstable.  Finally, they can incorporate covariances and other parameters derived from turbulent market periods rather than normally tranquil markets to construct portfolios.