Television's "Mad Men" has revived interest in clothing styles of decades past. Men sport hats, and women, gloves.

Similarly, market action over the last several years has revived interest in an option strategy: collars. Those who need a review of the collar strategy can find it in the archived Options 101 article from November 28, 2008. Back in February, in Trader's Corner articles published February 13 and 27, I detailed a number of studies comparing returns on long equity positions with and without collars.

All the studies examined in the February Trader's Corner articles produced the expected results: collars performed best in downturns while muting upside participation during strong rallies. Why are these results the "expected" ones? The collar is constructed by the selling of a covered call against a long stock, ETF, further-out long option or LEAP position, among other possibilities, with the premium from the sold call financing the purchase an OTM put. This construction caps the participation in upside gains because no further gains will occur once the underlying's price moves through the upside breakeven of the equity plus sold call position. At the same time, the long put places a floor under the equity's price, so that losses are stopped at a predetermined level. While upside gains during rallies might be muted, over the volatile period that has covered the dot-com bust, the subsequent rally, and the credit crunch, many collaring strategies have routinely outperformed a long equity strategy. At the same time, most collaring strategies have performed these feats while also reducing volatility in gains.

Now comes a new study by Edward Szado and Thomas Schneeweis, "Loosening Your Collar: Alternative Implementations of QQQ Collars," that covered a 122-month period from March 1999 to May 2009. March 1999 was notable for the introduction of options on the Powershares QQQ (QQQQ).

Szado and Schneeweis examined several strategies, including a passive collaring strategy and an active one. The active one required examination of momentum, volatility and macroeconomic signals. That strategy is probably best studied in depth by experienced options traders who go directly to the study's report to seek information from the study's originators.

One signal employed in that active strategy is interesting enough to merit mention for the benefit of experienced options traders, perhaps prodding them to go to the study. When looking at the volatility signal, Szado and Schneeweis, like researchers Renicker and Mallick before them, determined that when the volatility measure being used registered high short-term anxiety levels, only 0.75 calls are written to each long index position. When the volatility measure being used registered low short-term anxiety levels, 1.25 calls are written to each long index position. What this does is sell fewer calls and limit upside risk when volatility measures indicate that markets may be ready for a hard bounce (with emphasis on "may") and sell more when volatility measures indicate that markets might be ready for a reversal or pullback after a rally. I had missed the earlier Renicker and Mallick study and hadn't previously known of any studies on collars that altered the number of calls sold in this manner. Interesting strategy, don't you think?

The passive strategy is the more familiar strategy of buying either a 1-month, 3-month or 6-month put and financing the purchase of that put with a sold call. For the purposes of this study, like Lentz's study mentioned in one of the February articles, Szado and Schneeweis found the best returns on a 1-month call/6-month put collar strategy. Of course, in this strategy, it may take several months of rolling into a new sold call to pay for the put's purchase. For the purposes of the study, Szado and Schneeweis sold calls 2 percent out of the money while Lentz had settled on an ATM sold call as returning the best results on the underlying he chose.

Szado and Schneeweis divided that 122-month period into three subperiods: March 1999 to September 2002, October 2002 to September 2007, and October 2007 to May 2009. The second of those two subperiods encompassed a strong rally phase for the equity markets. The passive collaring strategy, as expected, underperformed the market during that strong rally, but still produced a 5.2 percent gain. Over the entire 122 months, the QQQQ returned an annualized -3.6 percent loss, while the passive collar position employed by the study returned a theoretical 9.3 percent gain. Remember to consider all such back-tested strategy results as theoretical. We don't know how much real-world trading realities might have impacted that strategy.

To see how the results shook out, their chart can be examined. The following chart compares a long equity position (dark blue line) with a passive collared position (red) and active collared position (light blue).

Growth of $100 in Active and Passive Collar Strategies from Szado and Schneeweis Study:

To find out more about the specifics of the strategies' implementation, particularly the more complicated active collaring strategy, the study results are available in a 46-page pdf available through the OIC. The study also details an alternative strategy for collaring a mutual fund by beta-weighting it and making other adjustments to find the appropriate collar.

I am not suggesting that subscribers employ these specific strategies. I am suggesting that collars are important strategies to study and then consider if one has long equity positions and capital preservation is of primary concern. This is particularly true if one holds long equity positions and either can not sell them or prefers a buy-and-hold strategy. Learn the pros of this strategy as well as the cons, which include transaction fees and a muting of gains during runaway rallies, but get to know them. Then, when you know enough, ask your money manager or broker about whether a collaring strategy is appropriate for your investments.