Volatility Gremlins

I had a conversation today and the effects of volatility over the long run was brought up. This is a concept that is very rarely talked about. I suspect not many people (advisors) have this discussion because they don't even know it exists. This topic is one of the main culprits individuals fail to realize the average returns over the long run that is to be expected based off of historical stock market return marketing material. The other culprits are inflation, timing, and making too many decisions. Historically, we have come to believe the stock market will return about 7.2% over the past 100 years or so. At issue is the creating of this figure. Using a geometrical average, which is how Roger Ibbotson who is credited for calculating this figure among others, you will arrive at roughly a 12% return figure. The geometrical average is calculated by simply adding up the number of data points and dividing by the number of data points used. For example the geometrical average of 2%, 5%, 15%, 20%, 7% = 49/5 =9.8%.

This work well as a marketing tool for long run performances. Brokers can always hide behind this widely accepted figure. This simple average is also used in calculating all kinds of forecasting models, pension funding requirements, and so on. However, a geometrical average is not what we experience in reality. Lets use the returns above as a quick example. Beginning with an account size of $100,000 the portfolio using the simple geometric average and return results as demonstrated above would gross $159,592.2. However, in reality you don't actually realize the average return each and every year. For every year you under perform the average you need a return in excess of the difference. Another words, if you start with 100% and lose 50% the next year you will have to have to make 100% to break even in the following year. In our example we saw two year of below average return followed by two years of greater than average returns. The gross return accounting for realized gains is $158,143.86, more than $1000 difference over 5 years. The difference here is relatively small but when you string together several years of under performance and maybe negative year or two and all of the sudden the difference becomes much larger. In the not to distant past, from 1/1966-1/1986 the market returned only 1.9% average. In reality does this volatility make any real difference? Over the same sampling period over the last 100 or so years the actual returns look something like 4.8%. Another words, over the long run, volatility erodes about 2.4% per year. Pretty significant and this is without even mentioning inflation, and when you begin your investment.

For a more detailed look at volatility issues I recommend reading, Bull's Eye Investing by John Mauldin.