The past few years have seen a marked drop in stock market volatility from the elevated levels seen during and in the aftermath of the Global Financial Crisis. Given the scale of the drop it is instructive to put recent experience into a longer historical context.

This context is useful for determining what, if any, investment implications follow from the recent fall in volatility and understanding the pitfalls of relying too heavily on recent market behaviour in making investment decisions.

Stock market volatility near 50-year lows

In Figure 1, we plot the volatility of the S&P 500 Index from July 1962 to October 2014. As the chart shows, recent volatility has been near the bottom of a 50-year range. This low is all the more striking because, in the immediate aftermath of the Global Financial Crisis, volatility was higher than at any time since at least 1960 and, in all likelihood, since the 1930s depression.

Figure 1: S&P 500 Index volatility estimated from July 1962 to October 2014

Dissecting index volatility

Changes in the volatility of indices such as the S&P 500 are driven by 1) the average volatility of constituent stocks; and 2) the correlation or dispersion between constituent stocks [1]. All else being equal, increased average volatility will increase index volatility, and increased dispersion will reduce index volatility. In Figure 2, we split out these two components.

Figure 2: Dispersion versus volatility of S&P 500 stocks

What drives changes in dispersion and volatility?

As Figure 2 shows, S&P 500 dispersion has varied greatly over the past 50 years or so. And, in the immediate aftermath of the Global Financial Crisis, this dispersion was lower than at any point since 1962, and would stay low for some time.

Changes in the underlying correlation structure are sometimes rapid and are therefore unlikely to mirror changes in the correlation of underlying fundamentals of the economy. Indeed, there is little to no relationship between changes in dispersion and changes in the correlation of indicators such as earnings.

Figure 2 shows how stock market volatility has also varied considerably over time, but its drivers are difficult to identify. Thus volatility is not tied to the volatility of macroeconomic variables such as inflation, money supply growth or industrial production [2]. If anything, the evidence points to stock market volatility driving macroeconomic volatility as policymakers respond to large moves in the stock market.

The “Greenspan Put” and quantitative easing are good recent examples of this [3]. There is, nonetheless, clear evidence of stock market volatility increasing in recessions and in times of uncertainty [4]. The shaded areas on Figure 2 show recessions as demarcated by the National Bureau of Economic Research.

Stock market volatility has risen sharply during most recessions since 1960, and has tended to fall once the recession is over and most of the local maxima in volatility occur in or immediately before contractions, as was the case of the dot-com bubble. Dispersion, by contrast, has tended to fall ahead of and during most recessions before increasing during the recovery phase.

There are many reasons why uncertainty might increase during a contraction. One good reason is that there are not that many contractionary periods and therefore less data to go on. In addition, the Global Financial Crisis was different from most recessions over the past 50 years because monetary policy was bounded by near-zero nominal rates ‒ new ground for anyone not trading during the Great Depression of the 1930s.

Recent volatility and dispersion in context

From this perspective, the past few years are fairly typical because the pattern of volatility rising and dispersion falling is one that has repeated itself several times over the past half century. What is unusual and striking about the recent period is the size of the moves in volatility and dispersion.

The recent increase in dispersion has not gone unnoticed by the active equity manager community. A claim often made is that the increase in dispersion improves the expected return of managers because stocks aren’t moving in lockstep.

But this isn’t correct. An increase in dispersion implies an increase in dispersion between different managers’ portfolios and thus likely implies increased dispersion across managers’ returns. Higher dispersion does not necessarily mean a “stock pickers market”. A “wheat from chaff market” would be a more apt description. Secondly, the current vogue for so called “smart beta” products is mostly based on simulated track records, and simulations that include the historically extreme moves we have seen in both dispersion and volatility. We have previously pointed out the risks of relying on such track records, and the long-term volatility and correlation data provide another reason for caution [5].

Conclusion

Looking at very long runs of data is informative about current conditions and is a good defence against short-run mistakes. We believe it is an essential part of the process of distinguishing between random and persistent features. Such distinctions are key to building robust trading systems in financial markets.

This article contains information sourced from S&P Dow Jones Indices LLC, its affiliates and third party licensors (“S&P"). S&P® is a registered trademark of Standard & Poor’s Financial Services LLC and Dow Jones® is a registered trademark of Dow Jones Trademark Holdings LLC. S&P make no representation, warranty or condition, express or implied, as to the ability of the index to accurately represent the asset class or market sector that it purports to represent and S&P shall have no liability for any errors, omissions or interruptions of any index or data. S&P does not sponsor, endorse or promote any Product mentioned in this material.