Option Collar as Portfolio Insurance

Yesterday, I looked at how to protect a portfolio from a big drop in stock prices using put options, but the result was unsatisfactory. Mark Wolfinger left a comment suggesting selling a call option as well to create what is called a “collar.”

This may sound complex, but the effect of a collar is simple enough. Your stock market returns are limited to a range. If stock losses are too deep, the purchased put options compensate you to limit your losses. If stock market returns are very high, the call options you sell limit your gains.

You may wonder why anyone would bother with the call option part if they limit your gains. The answer is that you get cash from selling the call options that you can use to buy the put options. So, in trade for the guarantee that you won’t lose too much, your gains are limited as well.

The particular strike prices of the options are what determine the limits on gains and losses. Of course, we prefer a tight limit on losses and a generous limit on gains, but we’d also like the cash generated from selling the call options to cover the purchase of the put options. This forces a balance between the upper and lower limits on returns.

An Example

Let’s continue with yesterday’s example of investing in large U.S. companies through the S&P 500 (which is now sitting at 907.00 as I write this). This time we’ll focus on limiting losses between now and December 2009.

If the S&P 500 were to drop to 800, this would be an 11.8% loss. If we don’t want to lose any more than this, we need to buy December 2009 put options on the S&P 500 struck at 800. The current ask price of these options is $55.60. So, we need to sell call options at the same price.

Unfortunately, options don’t come in enough choices to exactly match this price. However, we can sell two types of call options to get the right blended price. The bid price on calls struck at 950 is $64.30, and for calls struck at 1000 the bid price is $44.90. With the right blend, the average price is $55.60, and the blended strike price is 972.4.

Unfortunately, this blend of call options means a cap on stock returns of only 7.2%. So, the returns over this 7.5-month period are limited to between a loss of 11.8% and a gain of 7.2%. This doesn’t seem like a very good deal. In fact, assuming that the average yearly stock return is 12% and the yearly standard deviation is 20%, the expected return from our collar works out to 1.3% over those 7.5 months.

That didn’t work out very well, but it was based on just one choice of option strike prices. The following chart gives a range of possible collars and the expected return from investing with that collar. To understand the chart, imagine any vertical line cutting through the chart. The blue line is the top of the collar, the red line is the bottom, and the green line is the expected return over 7.5 months.



To decide if any of these collars are worthwhile, we need something to compare them to. Suppose that the alternative to a collar is just using a 70/30 asset allocation between stocks and a fixed income investment returning 2%. If we expect 12% from stocks, we expect 9% yearly from this allocation, or about 5.5% over 7.5 months.

To get an expected return of 5.5%, the collar ranges from a loss of 34% to a gain of 24%. This is definitely not very useful. If an investor is looking for protection, capping losses as 34% over only 7.5 months isn’t much protection. But, limiting losses further cuts into expected gains and makes this approach less desirable than simply buying some fixed-income investments.

Note that this discussion has ignored the commissions and taxes resulting from trading options.

Overall, I’ve still failed to find an option-based strategy for portfolio protection that competes with asset allocation. Perhaps the problem is that recent stock market volatility has changed option prices in an unfavourable way for collars. In any case, I’m not going to be trying any option strategies for portfolio protection until the numbers make sense.