I read two articles on options yesterday.  I am not going to link to them.  Both authors generally have good information.  But for some reason, each made statements I think need to be corrected for the sake of you having accurate information.  [For what it's worth, I've gotten advice in the past to let things like this go uncorrected, because it's in my economic interest to have uninformed masses believing things that aren't true, thus giving informed traders an edge.  What can I say, I'm irrational!]

One article was about how options forecast the expected magnitude an event will have on a stock price.  For instance, let’s say a stock is at 30, and the 30 call is 1.50 and the 30 put is also 1.50; the straddle is priced at 3.  The breakevens on the straddle are 27 and 33: 10% above the current price and 10% below the current price.

Based on that, some say that the options market is forecasting a 10% price swing.  Well, sort of.  It’s not wrong; it’s just incomplete.

Here’s why.  Straddles have an asymmetric risk/reward with uneven payoffs.  These are not binary options.  It’s not a game where you double your money if the move is larger than 10% and lose all your money if the move is smaller than 10%.  If you buy a straddle, you can make much more than you risk.  And if you lose, there’s only a very small chance that you’ll lose the entire investment.

The result is that the options market isn’t just forecasting a certain size event over a period of time.  It’s forecasting a certain probability of a 10% price swing over a certain period of time.  Not only that, the options are also pricing in a certain probability of a 5% move and a 20% move over a certain period of time.  For instance, the options market might be pricing in a 40% probability of a 10% move in 7 days.  Or a 5% chance of a 20% move in a month.

The point is, it’s not just that the options are pricing in a certain size move over a certain period of time.  There is an additional element.  With options, risk and reward always lead to a three-factor assessment: how big, how long, and how probable.

The other thing I read dealt with the low level of VIX.  It got down towards 17.  The article, which has been reposted on several major financial sites, stated that the low VIX signaled complacency.  I won’t dispute that.  In fact, I know I’ve said the same thing … in the past.

Where I have a problem is that the article then stated a causality: the low VIX and high degree of complacency is a bad omen for the stock market.  The question then becomes, is the low VIX really a bad omen for the stock market?

NO.

At least not when measured over the actual time frame of the options that VIX measures.  Remember, VIX measures constant 30-calendar-day implied volatility.  VIX makes no measurements of anything beyond that point.

A simple analysis of the numbers can tell us whether a low VIX is a bad omen over the next 30 calendar days.  Looking at the performance of the stock market 30 days after VIX is at 17.61, we find that the maximum drawdown of the market is much lower than when VIX is higher.  We find that the higher the VIX, the larger the drawdown in the market is over the following 30 calendar days.  The lower the VIX, the smaller the drawdown in the market over the following 30 calendar days.

This first graph shows the maximum drawdown of the S&P 500 when the VIX is 17.61 (last night’s closing level) or less.  You can see that the worst drawdown over a one-week period is about -7%.  The worst drawdown over a one-month period is about -10.5%.

 

Now let’s look at what happens at a different VIX threshold, say 25%.  The white line in the graph below shows what happens when VIX is at that level.  The one-week maximum drawdown is a massive -28%.  The one-month drawdown is -31.5%.  [The orange line in the graph below shows what happens when VIX is at 17.61%.]

Clearly, the white line shows that drawdowns are far more severe when VIX is at 25% than 17.61%.  What that means is, instead of a low VIX being a bad omen for the market, it’s actually been a good omen, at least with respect to drawdowns.  The conclusion in the article is just plain false.

Unsurprisingly, when VIX is high, not only are the drawdowns large, but the rallies are quite large too.  When VIX is low, the drawdowns over the subsequent month tend to be small, but the rallies are relatively small compared to when VIX is high.

All of this leads to what is probably one of the most obvious statements in the history of options: when VIX is high, the market is volatile; when VIX is low, the market is less volatile.

In other words, the low VIX is merely confirming what we already know.  The market just isn’t that volatile.  And it’s unlikely that volatility will suddenly explode to the downside in the next 30 days.

Here’s the caveat.  This analysis is based on the action over the past 23 years (I used VXO for the period prior to the current iteration of VIX).  Things can and do change, so this analysis doesn’t rule out a major decline.  Anything is possible.  But it sure would be unprecedented.

– Don

This entry was posted on Wednesday, December 15th, 2010 at 12:28 pm and is filed under Magnitude, Market Updates, Options, Probability, Volatility. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.