The prices at which options on various stocks and ETFs trade provide a great deal of insight. Back in November of 2008, I noted that options on a range of defensive stocks were highly over-priced relative to the price of options on the S&P 500. In early 2007, I noted that options prices were signaling a massive increase in volatility in major asset classes. The prices of options have provided remarkably consistent and reliable signals through the very volatility market conditions of 2008-2009.
The key variable in options prices is implied volatility. The price of an option is determined by the following variables:
- The price of the underlying stock or ETF
- The strike price of the option
- The dividend yield on underlying
- The expiration date of the option
- The expected future volatility of the underlying
With the exception of #5, all of these variables are directly observable. Because of this, we can take the price at which an option is trading and the first four of the variables above, and figure out the expected future volatility implied by the option prices—and this is the implied volatility. A very easy to use (free) calculator is available at iVolatility.com. This calculator provides all of the inputs—all you have to do is to specify the expiration dates of the options you want to examine.
In August of 2007, I wrote an article in which I showed that returns on just about every asset class were negatively correlated with changes in implied volatility. With the options markets projecting that volatility would rise substantially, this implied that the prices of every major asset class were due to come down. Rising implied volatility signals declining prices on the underlying indexes and vice versa.
By November of 2008, market volatility was enormously high, but the long-dated options projected much lower volatilities in the future (see article in Footnote 1). This signaled that implied volatility would drop—and it has for every major asset class. Furthermore, there was a massive disconnect in the market when we compared the implied volatility on the S&P500 to that of a range of individual stocks. In particular, if we calibrated Quantext Portfolio Planner (QPP) to agree with the implied volatility on SPY and then looked at projected volatilities for a series of individual stocks, the projected volatilities of the individual stocks were much lower than the implied volatilities for those stocks. If we believed QPP’s projections, the market was irrationally risk averse with respect to these stocks. The way to profit from this irrationality was to sell options (which I did at the time).
As promised in that original article from November of 2008, I am revisiting this strategy to show how things have gone. In the original article, I compiled the bid prices of call options for an investor who might want to exploit the mis-pricing by selling call options:
click to enlarge
Options prices from November 2008 (from original article)
Options on these stocks were far too expensive relative to the prices of options on the S&P500 (SPY) according to QPP.
The Jan 2010 call option on Johnson and Johnson (JNJ) with a strike price of $65 could be sold in November 2008 for a price of $5.50. Today, that option is priced at $0.60. If you had sold the call option, you would be up by $4.90 per share today ($5.50 - $0.60). At the time, JNJ was trading at $58.22 per share (see table above). Today, JNJ is trading at $55.35 per share. An investor who owned JNJ is down by about 5% since then. An investor who owned JNJ and sold the call option is up by 3.9% over this period.
As another example, we can consider Southern Company (SO). The Jan 2010 call option with a strike of $40 could be sold at $2.60 per share in November 2008. Today, that option is worth $0.10. The share price of SO has gone from $35.22 to $31.17 as of this writing. The investor who owned SO is down by 11.3% over this period. The investor who owned SO and sold the call option is down by 4.2%.
The Jan-2010 $70 call on PG that was trading at $6.70 in November 2008 now trades at $0.15.
The Jan-2010 $70 call on KMB that was trading at $3.60 in November 2008 now trades at $0.10.
The drop in options prices has been precipitous and investors who were willing to sell these options into an irrationally risk averse market have profited substantially (the title of the original article is Profiting From Risk Aversion).
It must be understood that options prices naturally drop through time due to time decay. The shorter the period until expiration, the less an option is worth. What tends to validate our strategy, however, is that the implied volatility of these options has dropped so substantially in time. The following table is from the November 2008 article:
Back then, the implied volatility of a Jan 2010 $65 call option (the call option expiring in January of 2010 with a strike price of $65) on JNJ had implied volatility of 38%. Quantext Portfolio Planner (QPP) projected that it should have been at 24%, given where options on SPY were trading. QPP was projecting that the implied volatility on JNJ was way too high. Today, the implied volatility of that option is around 22%, which tends to validate QPP’s projection.
Similarly, we can look at the options on SO. The implied volatility on the Jan 2010 $40 call option on SO was 37% but QPP projected that it should be at about 20%, given where options on SPY were trading. Today, the implied volatility for this option is 22.7%.
The implied volatility on the Jan 2010 $70 call option on PG was 46% in November 2008. QPP projected that it should have been 22%. Today, the implied volatility on that option is 25%.
The most extreme case of relative mis-valuation identified in the original article was KMB. The implied volatility of the Jan 2010 $70 option was 45% but QPP projected that it should only be 11%. Today, the implied volatility of this option is 22.5%.
Note: the option prices and implied volatilities cited here are as of June 23, 2009
These results show that QPP correctly identified a substantial anomaly in implied volatility, caused by the extreme risk aversion of late 2008. The rational connection between implied volatilities (risk) for a series of individual stocks and the implied volatility of the broader market broke down. The options prices were suggesting that a series of companies were very risky. Today, much of this anomaly has corrected itself (and this will be discussed in a follow-up article). The very high implied volatilities in the individual companies listed above suggested that these firms had substantial risk of going under, because implied volatility is related to default risk. When I looked at QPP’s projections in late 2008 for the volatilities of companies like JNJ and compared these to the implied volatilities, I was willing to bet the substantial default risk priced in to the very high implied volatilities did not make a lot of sense.
The results of this strategy have also benefited from a reversion of implied volatility levels in the broader market to more rational levels. I identified a series of stocks for which the implied volatilities were irrationally high, given that implied volatility on SPY was 45%. There has never been an extended period during which volatility on the broader market was this high, so it was also rational to bet that this would drop. Implied volatility for long-dated options on SPY is now below 30% as is the VIX, which measures implied volatility for options expiring in the near-term.
In my next article, I will provide an overview of what this method for valuing options relative to the implied volatility on SPY is telling us now.
Note: all calculations shown in this article were generated using Quantext Portfolio Planner (QPP) will all default settings
Disclosure: the author is long JNJ and SO and long-dated calls on JNJ and SO as of this writing.