The importance of this becomes evident when we consider how this fact can help keep portfolio risk levels constant throughout the investment horizon. If indeed today’s volatility predicts tomorrow’s volatility, and today’s volatility exceeds our predetermined risk levels then we can easily adjust our risk by reducing our position sizes dynamically.
We illustrate this concept in the chart below.
The lighter blue line represents the returns in the S&P 500 if we had bought and held it from 2004 to 2011. The darker line is also an investment in the S&P 500 but here we systematically add a larger cash component to the portfolio as daily volatility increases. In fact, our goal is to keep our daily volatility levels the same, regardless of market conditions, at no more than one percent a day or 15 percent a year. Changes in volatility are depicted by the heat bar at the bottom, the brighter the red the higher the volatility and consequently the larger the cash component.
What we observe is that this simple concept of sizing positions based on yesterday’s volatility is enough to outperform buy-and-hold by 125 percent while also keeping volatility constant at around one percent a day, even during October of 2008 where S&P 500 volatility went as high as 7.11 percent.
To illustrate further, this graph highlights some of the major events leading to the credit crisis and the corresponding cash position required when using this basic approach. As we can see, large cash positions of up to 84 percent were necessary in order to keep daily volatility constant at our target of one percent.
Source: Yahoo Finance, DarwinFunds.ca
Source: Yahoo Finance, DarwinFunds.ca
What is also obvious is that this type of risk management not only helped keep portfolio risk consistent but also aided quite substantially in providing better absolute and risk-adjusted returns over time.
To further illustrate the relationship between returns and volatility, we compare market compound returns during periods of high volatility versus periods of low volatility. In the bar graph below, we divide the volatility of the S&P 500 into quartiles from highest to lowest and assign them to their corresponding market return.
What is also obvious is that this type of risk management not only helped keep portfolio risk consistent but also aided quite substantially in providing better absolute and risk-adjusted returns over time.
To further illustrate the relationship between returns and volatility, we compare market compound returns during periods of high volatility versus periods of low volatility. In the bar graph below, we divide the volatility of the S&P 500 into quartiles from highest to lowest and assign them to their corresponding market return.
We can clearly observe how markets seem to perform best during the periods that are more predictable and less volatile. It then follows that we should focus on reducing our allocations to asset classes that are exhibiting erratic behaviour and increasing our allocations to those asset classes that exhibiting calmer and more predictable behaviour. This is a more optimal portfolio management approach given that it works across all types of asset classes as we can see from the graphs below.
Volatility analysis is the cornerstone of our portfolio management process and the one that is the most likely to persist through time. When used in combination with correlation and momentum analysis, it creates one of the most robust and consistent wealth management tools available to us.
