Collaring multiple-asset ETFs to manage tail risk

July 31, 2012 07:00 PM

Since the 2008-09 financial crisis and the central bank’s response of near zero interest rates, financial markets have been vacillating violently between risk-on and risk-off periods. This bipolar attitude toward risk has increased asset class correlations, negating the effective benefit of many traditional equity diversifiers. In this unpredictable climate, investors are focused on seeking strategies that offer downside protection but also upside participation.

One solution to achieving both of these often mutually exclusive desires is an option-based equity collar. A long collar strategy involves owning the underlying, while being long a put option with a strike price below the market and short a call option with a strike price above the market. This orientation of the strike prices makes each put and call option out-of-the-money (OTM). An option collar can provide portfolios with greater downside risk protection than standard multi-asset diversification programs, but they also allow for profits during risk-on rallies.

Recent research has examined the performance of the collar strategy against a range of exchange-traded funds (ETFs) across multiple asset classes, including equity, currency, commodity, fixed income and real estate. The resulting book, “Option-Based Risk Management in a Multi-Asset World,” was authored by Research Analyst Edward Szado and University of Massachusetts Isenberg School of Management Professor of Finance Thomas Schneeweis. Their analysis shows that for most of the asset classes considered, an option-based collar strategy, using six-month put purchases and consecutive one-month call writes, provides a holy grail of investing of improved risk-adjusted performance and significant risk reduction. 

“Collar growth” (below) illustrates the benefit of an equity collar strategy on the popular SPDR S&P 500 (SPY) ETF. Over the 55-month study period ending Dec. 30, 2011, the 2% OTM passive SPY collar returned more than 22% (4.5% annually), while the long SPY experienced a loss of more than 9% (–2.1% annually). The collar earns its superior returns with less than half the risk as measured by the standard deviation (8.4% for the collar vs. 19.5% for SPY). 

One of the most telling statistics supporting the potential benefit of equity collar protection is the maximum drawdown. During the study period, SPY experienced a maximum loss of 50.8% while the 2% OTM collar reduced this negative performance by four-fifths to a maximum loss of 11.1%.

Trading approach

In their book, Szado and Schneeweis evaluated the impact of collar strategies across a wide range of asset classes based on a set of trading rules (see “Cross-market analysis,” below).

For each of the ETFs, the study analyzes an array of strike prices with defined amounts of OTM values where a six-month put is purchased and consecutive one-month calls are written. At the close on the day before the Saturday expiration of the calls, and depending on the particular passive implementation, the initial amounts the calls and puts were out-of-the-money were set at: 25%, 10%, 5% or 2% or at-the-money (ATM).

At expiration, the calls were settled at intrinsic value, and new one-month calls with the specified amount of OTM were sold. The longer-term put is held for another month. Once the six-month put expires, it is settled at intrinsic value and again rolled into new puts and calls with the specified amount of OTM and time to expiration. 

A rule was created to avoid circumstances in which the underlying prices declined significantly and new calls then would have been written at lower (crossed) strike prices than the existing deep in-the-money (ITM) puts. In these cases, long put and long underlying positions would counteract each other and the new call essentially would be written naked.

Because writing naked calls is inconsistent with the risk-reduction purpose of the collar strategy, Szado and Schneeweis implemented a rule to roll the puts to the strategy’s target amount of OTM and maturity based on the current underlying price on the day the new short call position was initiated. The put sale was rolled at the mid-point between the bid and ask, and the new put was purchased at the ask price. All other executions in the study include transaction costs by purchasing at the offer price and selling at the bid price.

Results analysis

In contrast to earlier studies that concentrated on equity markets, Szado and Schneeweis provide extensive analysis of the performance of collar strategies over a diverse set of asset classes including equities, currencies, commodities, fixed income and real estate.

Their study covers the period from June 1, 2007, to Dec. 30, 2011, except for the gold-based ETF, which begins on July 1, 2008, the first full month after the June 20, 2008, inception of gold ETF (GLD) option trading. The period of study was chosen to capture both the financial crisis and the subsequent market recovery. The risk-reduction and drawdown protection ability is evident in the results across all asset classes analyzed. 

As “Cross-market analysis” shows, the results are somewhat mixed from the perspective of total-return qualities. Two currencies (the Australian dollar and Japanese yen), the two bond ETFs (HYG and TLT) and the Nasdaq ETF (QQQ) and GLD had higher cumulative returns than any of the collar iterations.

In general, the analysis suggests that with respect to total returns, option-based collar strategies tend to outperform when drawdowns are more aggressive and tend to underperform in periods of extreme run-ups. 

More importantly, while option-based collars may not provide complete protection or maximum returns for all products and in all market conditions, collars can provide significant risk control across a wide range of asset classes, significantly reducing volatility and drawdowns. In certain market environments, the collar strategy also can provide enhanced returns relative to a stand-alone investment.

Philip Gocke is managing director of research and educational strategy for institutional investors at the Options Industry Council. He is responsible for educating pension funds, hedge funds and money management firms about the benefits and risks of exchange-traded options. He has worked for Van Der Moolen Options U.S.A. LLC, Bank of America’s Financial Market Advisory Group and the Federal Reserve Bank of New York.

About the Author
Phil Gocke is managing director of The Options Industry Council. The Options Industry Council (OIC), which provides education and research to institutional investors, helped sponsor the CISDM study of the performance of a collar strategy on the PowerShares QQQ exchange-traded fund (ETF).