In recent months volatility has taken a breather on Wall Street, particularly in the last month, as stock market volatility has virtually come to a stand-still. Below is a chart showing the daily changing closes on the S&P 500 since the beginning of August. We see in the chart that except for a few days, since mid-summer, there has been almost no substantial variation in day to day changes. Over that time, the daily median change (up or down) has been an incredibly low .30%. The daily standard deviation over this time has been about .60%, equating to an annualized yearly standard deviation of about 9.5. According to data from Yahoo Finance and the CFA Institute, the daily volatility in this most recent three-month period is about 60% of average daily variation since 1950.
According to Crestmore Research, since 1950, there has been an average of about four days per month (a month averages about 21 trading days) in which the daily close in the U.S. stock market (as measured by the S&P 500 Index) from the previous day was up or down more than 1.0%. From the low volatility side of the spectrum, in the mid-1960s, there were multiple months in which the stock market did not change 1% from the prior day during the entire month. Conversely, in the not-too-distant past, as financial panic hit Wall Street, there were months where half of the days of the month saw 1% changes.
Looking at the last three months, in August there were no 1% changing days, in September there were five, and in October, there was just one. That’s a total of six 1% daily market changes in three months, or an average of just two per month. Daily intraday volatility has also been extremely low with just 14 of the 65 trading days experiencing a 1% range between high and low. In times of normal volatility, it is very common for the majority of days to have a greater than 1% trading range.
The next graph covers the same August to October period, but from 2008, on the eve of the financial panic. For comparison’s sake, I used the same volatility scale so that it is easily comparable to the 2016 version of the chart.
We see that even before the panic really took hold in the fall, volatility was already much higher than today. The volatility in both August and September of 2008 was more than three times that of their respective months in 2016. The high October 2008 volatility compared to the very low October volatility of this year, resulted in a volatility measure more than 13 times higher than today’s recent volatility. To put such daily changes in perspective, the October 15, 2008 one-day decline of 9.84% would be equivalent to nearly 1,800 points on the Dow Jones Industrial Average. The cumulative October 2 to October 10 decline of 2008 would equate to about a 4,800 point decline on the Dow Jones. Such a decline is a reminder of the potential levels of volatility that are always inherent in the market.
Looking at recent month-to-month (month-ending) volatility, we see much the same story: very low volatility. Below I’ve compared monthly changes over the last seven months to that of the same periods in 2008 and 2012. The difference in monthly volatility between 2008 closes (in blue) and those in 2016 (in green) is stark, with average monthly changes (up or down, absolute values) more than five times higher in 2008 compared to this year. And while much of 2008 was abnormally volatile, I also added 2012 (a year I chose at random from the six years in-between). We still see that monthly volatility during these months in 2012 averaged more than twice as high as in this year.
By the way, the longer-term period leading into the problems and volatility of 2008 was a time of low volatility, comparable to levels of this year. One market tendency is that bull markets are accompanied by declining and low volatility, followed by increasing volatility and subsequent market collapse. This was the case before the 1973 bear market, the 1987 crash, the 2000 collapse and the 2008 panic. Coupled with today’s historically high valuation - comparable to 2000 and 2008 – increasing volatility in the months ahead will likely be a sign that investors should buckle down for the possibility of a substantial market correction.
At the very least, today’s very low volatility will soon be reversed with more typical volatility. Or perhaps more likely, today’s low volatility might soon be replaced with higher-than-average volatility, such as existed in 2008. That’s because another market tendency is for the market to swing to extremes. Like about everything else in the stock market, volatility swings like a pendulum, ranging from cyclical periods of low volatility to periods of higher volatility.
Even by the standards of relatively low volatility over the last six years or so, the current period of low volatility over the last few months has been remarkable (see graph below). We see from the chart that the last comparable period of very stable stock prices lasting several months was in mid-2015, right before a significant correction and period of increased volatility over the following several months.
It is impossible to know all the reasons why there has been a temporary lull in stock market volatility. Obviously, this summer and fall there has been tremendous focus on the U.S. Presidential election. This has pulled much attention away from problems with global economies, conflicts and wars. In addition, the price of oil has lately been in a “sweet spot” for investors, not too high and not low. That is in contrast to periods of high oil prices, and accompanying inflation, which is always stressful to investors or periods like the beginning of this year where extremely low oil prices were weakening the stock market, because of the stress low oil prices were putting on oil companies and the high-yield bond market.
There is also a perspective that financial markets, in the U.S. and outside of it, are more “managed” than ever by governments and central banks. It is no secret that the Federal Reserve is constantly intervening in various financial markets providing “liquidity” (buying and selling various types of assets) or taking it away, depending on market conditions. Managing interest rates is another method the Fed uses to control the markets. The Fed’s pulling back intentions to raise interest rates this year despite stronger economic numbers than a year ago has also helped to keep financial markets levitated.
Regardless of the cause of recently low market volatility, such a period will end. This will perhaps come with a growing increase in volatility with stock prices steadily rising or falling. Or perhaps volatility will spike when stocks crash.
Investors that don’t have a tolerance for high market volatility should be realistic as to whether or not they are willing to bear the risk and inherent market volatility, even when, as of late, it is sometimes hidden for a time. An exercise that I think is good for investors to practice is running through possible scenarios of potential daily volatility and downside risk.
So that might mean considering an event where the stock market falls 3% in a day or 10% in a month. Or it might mean a stock market correction of 20%, 30% or 50%. Investors should always make it personal to their circumstances. If I have a $100,000 stock portfolio and the market goes down 10% in a month, that might mean a decline of $10,000 of my investments. Am I willing and able to accept that risk? How about seeing my $100,000 stock portfolio get cut in half over a year or two? Can I stomach that risk? If not, perhaps I should be looking for a way to lessen that risk before such an event.
The research into the volatility of the financial markets has shown quite clearly for some time that the stock market is more volatile on the extreme end of the scale than history and mathematics would predict. The phrase “fat tails” has been used to describe the tendency for extreme market events to be more numerous than statistics and a normal distribution would predict. Interestingly, analysts also see that the market during non-volatile times, is actually less volatile than one would expect. In other words, markets are not very volatile until they are. Then they’re even more volatile than they should be. In the financial crisis of 2008 there were multiple examples where markets and correlations basically did things that a mathematician would have considered nearly “impossible”, even over millions of years of financial market activity.
The CFA Institute (cfainstitute.org) has done some interesting research on stock market volatility. Based on a normal distribution of stock market activity, a “six-sigma” event should only occur once in about every 10,000 years of trading, but research shows that for the last several decades these six-sigma have on average been occurring every two or three years. Analysts are also finding that not only are extreme markets events more common than they should be, but over the last several decades, the levels of such extreme activity has been steadily rising. So in the 1950s the number of “four-sigma” (four standard deviations or greater) events happened only about once every 1200 trading days or so, that number had increased to about one in every 250 days by the 1980s, and to about one in every 50 days by this century. Clearly markets don’t always do what they’re “supposed” to do. The possible reasons for such changes are varied and complex, but the fact remains that despite all the times that investors would like to believe otherwise, market volatility has not been erased from Wall Street.