What We Can Learn from Volatility History

There are no short cuts to easy profits, but sometimes Mr. Market throws us a bone (or two) in our quest for strategic insight. Two examples, though hardly the only ones, come by way of reviewing correlation and volatility histories. We surveyed correlations last week, and today we revisit volatility, as per our chart below, which graphs rolling three-year volatilities of monthly total returns back to September 1994.

The obvious trend is that there is one, or so it appears. Lulls in vol tend to be followed by surges, and then lulls, and round and round we go. It's all obvious in hindsight, of course, but dissecting where we are in real time and how long it will last (or not) is always more obscure.

Meanwhile, we're constantly fighting our own biases. That's partly because the mind likes to extrapolate recent activity far into the future. Back in the late 1990s, the calm in volatility readings was thought to be the dawn of a new era in smooth and high returns in stocks and other risky assets. Such thinking prevailed right up until the idea was beheaded in the crash of 2000-2002, a reversal of fortunes that was accompanied by a spike in vol.

Something similar unfolded during 2003-2007, when returns generally were strong and standard deviations were low. Once again, it was all too easy to believe that the trend would last. It didn't, leading to a collapse in returns and a swelling in vol.

It's always hazardous to speculate on when a current cycle will end and a new one will begin. Nonetheless, we can and should observe cycles for what they are: finite. The one that now has the world by the neck may appear set to roll on indefinitely, but that is an illusion. This too shall pass, although the timing, as always, is unclear.

We can, however, search for some perspective, and to some extent we can find it. Granted, it's thin and subject to revision on a moment's notice. As such, we must be wary, even when it seems as though we've stumbled upon something meaningful. With that caveat in mind, consider our second chart below, which can be read in context with the first graph above. Note that relative high performance tends to be accompanied with relatively low volatility. No, it's not a perfect match, but there's a general rhyme here, or so it looks to our eyes. Similar trends pop up over longer histories as well.

In fact, we suggested as much back in early 2007, noting at the time that the lows in vol looked worrisome. One reason for our concern then was the history of blissful marriage of low vol and high returns ending with a spike in volatility and lower if not negative returns. Does it always unfold like that? No, but it has occurred enough over the decades to keep us wary, and watching the trends.

Keep in mind that in both charts we've smoothed the data, i.e., the lines reflect trailing 3-year trends. As such, none of this is much help to the day trader and it's only of limited help, if any, to strategic-minded investors. Nonetheless, the graphs suggest that the current rise in volatility is still quite young, or so it appears next to the previous stumble in 2000-2002.

One could argue that volatility is still rising and returns are still falling. Indeed, trailing 3-year returns, although they've fallen sharply this year, are more or less flat. By comparison with the previous cycle, can we expect some red ink in trailing three-year returns and perhaps some higher volatility yet? Let's not rule it out. Even if vol has reached its highs this time around, it wouldn't be out of character for standard deviations to persist for a time.

Considering all this, one could reason that the correction still has a ways to go. Again, that's pure speculation and so it must be taken with a grain of salt. What's more, volatility and correlation offer only two perspectives, and by themselves they're of limited value, which is why we also review a number of other metrics, including dividend yields, economic conditions, and so on.

Summarizing our research, only of which a portion appears on these pages, we expect that the correction will roll on, although we're generalizing broadly in terms of asset classes. Some look better than others at this point, and that informs our asset allocation. But to the extent we're making general observations, volatility is just one reason. The economic outlook weighs heavily on our thinking, too. And for the moment, it's unclear just how much blowback is coming to Main Street.

Comments

[JasonC:]

Um, high volatilities appear to persist in 3 year trailing averages for average periods of, you guessed it, about 3 years. Duh. Even if the periods of high volatility were short, that would be true, as long as they contribute most of the deviation in vol from the long run average.

Yes, higher vols correlate with lower returns. You don't simply see only large down moves - moves both ways get larger on a vol spike - but the next movement in the high vol period is usually down to flat. While the net move in low vol period is usually up, and substantially so.

The one item in the article that is most correct is that high vol periods are finite and they fade. Volatility does not follow a random walk - its dispersion does not grow as the square root of the time window, for instance. Instead volality is mean reverting. High vols tend to drop back to normal on a quite short time scale.

There are two ways to use this, one as an indicator for stock investing, and the second directly. When vols are low but haven't been for long, it is a good time to invest. When vols are low but have been for years, that is the danger signal. When vols rise, protect initially, and get greedy if the volatility persists and prices have already fallen.

The more direct way to play it is simply to trade the vols themselves as mean reverting. That means you want to be long option premium when vols are low and options are cheap, and short option premium when vols are already on a spike and options are overpriced.

If you couple the two, you get a strategy like this -

hold stocks long term.
buy puts to insure them only when vols are low and have been for 1-3 years.
once vols spike, deliver against your puts.
sell premium both directions during the vol spike (short straddles e.g. - you can cover with far out of the money options to control risk etc)
re-establish long stock positions after stocks drop, while vols are still high. You can sell covered calls on them above market in that period, too.
Once vols drop back to normal, just hold stock long, no options used, for a year or two.
Then go back to using some puts as insurance etc.

FWIW.

2008 Oct 20 02:32 PM

[JasonC:]

As for the other fellow's comment, the whole point of having options and using them is we do not need to know what will happen next to make money. It is enough to know whether a lot will, or whether people's fears of that have reached the "overblown" point.

Which VIX at 70, they have clearly reached the "overblown" point right now. With stocks already down 40% from their peaks, it is likewise already clear they are cheap enough to re-establish long positions. If you want, sell the VIX-inflated calls above market as you do so. Or sell straddles with half your capital, hold stock long with the other half.

(It should go without saying, but an option position to a certain amount of capital means the principle amount in bonds earning and as your collateral, and net option bets only the size of your interest, not your principle).

2008 Oct 20 02:35 PM

[stocks:]

nice article keep providing a good information everyday

2008 Oct 20 02:46 PM

[aarc:]

What do you expect volatility will be in a C wave?

If you look at the DOW and SnP500 on the yearly chart; they are defining a C wave down. The A wave down was from year 2000-2002, B wave up from 2002-2007. Now the C wave down should consume less time than the A wave and usually is much more volatile than the A wave. Problem is, nobody can really predict where the C wave will end until it is over and done with. It can either be a running correction C wave or an expanded flat C wave for the Dow or a C wave with higher low or a C wave with lower low for SnP500. DAX has the same double top pattern as SnP500 on the yearly chart and is now at the important level of 62.8% fibo retracement level to 2002 level.

It does'nt matter anyway for long term investment whether DOW will go to expanded flat target of 4950 or stop with a running flat level of 8000-7000 level. I prefer the running flat since an expanded flat on the monthly flat can be a killer; just look at MER monthly expanded flat, MER was able to complete the pattern but MER got consumed by BAC in the process. Likewise, DOW 7000-8000 levels is much more preferable than DOW 4950. If something really bad happens from now to first half of 2009, DOW, or for that matter, capitalism might not be able to survive massive shocks.

The important part of the yearly chart is that during the great depression in which the DOW went down from the high of 400+ to 42; there was a sustained rally from 1932 to year 2000 for a total of 68 years interrupted only by minor corrections. For Dow, 2007 level of 14000+ was not the top of the rally, 14000 was a part of the expanded flat or running flat correction that started in 2000. That is why SnP500 and DAX double tops are more important to chart analysis than DOW. Volatility in the 1932 went up to 186 if I have my memory correct. We are now at 80. Will it go to 186 or more or be less than that this time around? Who knows?

What is more important again as far as long-term investment is involved?

Well, we could be facing another 68 years of sustained rally interrupted by minor corrections again once this 8+ years of correction is done. It is not so often that we have this kind of massive stock market upheaval all of us less than 70 years old have never seen before and may not see again in our lifetime.

That is the important part.

2008 Oct 20 03:13 PM

[Gerry Sullivan:]

The "minor correction" of 1973-1974 was more severe than the current bear market (so far). I think that bear market can be viewed as a generational event lasting from Feb 1966 until Aug 1982.

In that 16.5 year span the Dow Jones Industrial average was down over
22%

2008 Oct 20 03:26 PM

[bobomite:]

How about mixing some economic observations with this pure "watch the lines" technical trading?

2008 Oct 20 04:10 PM

[aarc:]

Economically, there are many similarities between 1930s and today. First and foremost was the so called Roaring 20s where too much speculation was invested in the industrialization and the housing markets. Ditto Technologization and the housing markets. But mostly, it was the housing markets that was blamed for the collapse.

Naturally, when everybody suffered from industrialization/tech... hiccups; everybody looked to housing as a safe bet. But later on, housing bubble do collapse too, not only business. Then everybody realizes that housing is not an investment per se for everybody but rather an expenditure (unless you are in real state business)

Look what happened after the technology implosion. The US invested $1.2T into the housing sector. Almost all other investment for business went to China and India and other developing countries.

The sooner America realizes that housing is an expenditure and business enterprice is profit-generating investment; the sooner America can go back to business entrepreneurship.

Business entrepreneurship is what drives the markets, not the fake housing investment mentallity. The economy cannot sustain growth by sustaining expenditure rather than profit generation.

Every businessman knows that only with a successful business will you be able to buy other businesses, several houses and lots, all the expensive cars you want, even a yatch or an airplane for that matter. Now if you invest in a house and lot; will it be able to generate monthly or annual profits in order for you to buy what you want? Well, Americans did try by borrowing against their house, and look what happened.

While the business people were shoving profits into "safe investments" such as treasuries and bonds in anticipation of hard times ahead; the common employees were spending like there is no tomorrow thinking that their house and lot prices will go up forever.

Now, the Treasury has become a giant bubble by itselt. When will it collapse is a matter of time.

That collapse is what we have to watch with interest. Once investors start withdrawing cash from treasuries, they will have to deploy that cash into profit generating investments such as manufacturing plants, technology, energy, etc. Question is where? Will it be the USA or developing world again.

The US has a definite advantage over Europe when it comes to financials. And with the sudden resurgence of megabanks such as JPM, BAC and WFC in addition to C. There are now 4 horsemen of the financial world with the capability to offer the whole world a one-stop banking and finance system.

Unless the US can invent another system comparable to the industrial revolution and/or the computer revolution; it will have to depend on financial revolution to usher the whole world in general and the developing countries in particular into a New World Order.

At least a New World Order as far as the stock market is concerned.

Remember, the stock market is the best place for most businesses to raise capital. So developing countries will have to fully develop their stock markets. But you need the financial firms in order to raise cash. Hense the 4 horsemen of financial world would be assisting most of the cash raising endeavors thruout the world specially in the developing countries who are not sophisticated enough in global finance and the emerging stock market globalization.

Did I say emerging stock market globalization? Unless you live under the rock, you would'nt realize that stock markets all over the industrialized world are now interconnected and an attempt in the late 1990s to form global brokerage system has some form of successes. What is lacking is global cooperation among the western leaders. Until lately, it has always been the private sector that kept on pushing the stock market globalization.

But now, the G7 leaders are pushing for the resolution of the "global" financial market credit crisis. What is their solution? Problem really is stock markets tanking from country A to country Z. Solution? Globalized the darned western stock market in order for the western world not only to be able to be able to system shocks such as the credit crisis but also to be able to compete effectively against China, Russia, India, Brazil and other developing countries in the emerging global stock market competition. Much like when the western world has to compete with each other economically through industrialization and technology after WWII. The underdevelop countries were the customers too.

Now, the western world cannot sell autos, computers, and other things to the developing countries because those very countries are selling those very things to the western world now. The underdeveleped world of the 19th century are now becoming the new Industrialized World. We call them developing world. Developing for what? Developing for industrialization, that's it.

The Western Industrialized World has to either become the Financial World/Leaders or the Alternative Energy World/Leaders since it is now being dethroned by the developing world in Industrialization with no solution to win that title back in sight.

But since the Alternative Energy is really not a viable alternative right now; the US and Europe have to settle with Finance. After all, credit is now the de-facto commodity in the western world - as important as rice to rural China or India. So the west will have to sell credit (instead of cars, computers, etc) or the financing thereoff to the developing countries which are still in the infancy stage of global creditization.

Hence, another 68 years rally is not far-fetch if we think outside the industrialization/tech... box.

2008 Oct 20 05:32 PM