Testing A Simple Index Covered Calls Strategy
Posted in Equity Options
November 17, 2011
Does iteratively selling short-term, slightly out-of-the-money covered calls on a broad stock market index reliably outperform buying and holding the index? In other words, does any premium associated with selling the options outweigh the costs of execution? Two ways to address this question are: (1) apply estimated trading frictions to the CBOE S&P 500 BuyWrite Index (BXM), a benchmark index designed to track the performance of a hypothetical buy-write strategy on the S&P 500 Index; and, (2) examine the relatively short record of PowerShares S&P 500 BuyWrite (PBP), an exchange-traded fund (ETF) implementing the BXM index. Using monthly (options expiration date) data for BXM, PBP and S&P 500 SPDR (SPY) as a benchmark from the inceptions of SPY (February 1993) and PBP (January 2008) through October 2011, we find that:
CBOE describes the BXM index as follows:
“Announced in April 2002, …[d]ata on [hypothetical] daily BXM prices now is available from June 30, 1986, to the present time… The BXM is a passive total return index based on (1) buying an S&P 500 stock index [SPX] portfolio, and (2) “writing” (or selling) the near-term S&P 500 Index “covered” call option, generally on the third Friday of each month. The SPX call written will have about one month remaining to expiration, with an exercise price just above the prevailing index level (i.e., slightly out of the money). The SPX call is held until expiration and cash settled, at which time a new one-month, near-the-money call is written.”
The “cash settled” process for options that expire in-the-money, per “Description of the CBOE S&P 500 BuyWrite Index”, is based on the actual difference between the S&P 500 index and the option strike price. CBOE also explicitly qualifies returns for the BXM index via the following disclaimer:
“The BXM Index is designed to represent a proposed hypothetical buy-write strategy. Like many passive indexes, the BXM Index does not take into account significant factors such as transaction costs and taxes and, because of factors such as these, many or most investors should be expected to underperform passive indexes. Transaction costs for a buy-write strategy such as the BXM could be significantly higher than transaction costs for a passive strategy of buying-and-holding stocks.”
The BXM index estimates option cost by using the average of actual trade prices of the selected option series during a specified short interval. This approach is optimistic since the underlying strategy is always selling and would therefore tend to accrue the bid rather than the average. To bring the hypothetical closer to reality, we assume the buy-write portfolio bears a monthly trading friction of 0.01% to 0.20% of its value to sell call options and implement any cash settlements at option expirations. Real-life frictions would depend on actual broker fees and the size of the portfolio.
The following chart shows the cumulative values of $1.00 initial investments in the gross BXM index as calculated by CBOE and SPY since the February 1993 option expiration date. Results indicate that BXM tends to fall behind during bull markets and catch up during bear markets, with terminal value ($4.05) close to that of SPY ($3.97).
The average monthly return for BXM (SPY) over this period is 0.69% (0.74%), with standard deviation of monthly returns 3.47% (4.88%).
Over the subperiod since BXM’s live introduction in January 2004, the average monthly return for BXM (SPY) is 0.35% (0.40%), with standard deviation of monthly returns 3.94% (5.41%).
How do estimated trading frictions affect the performance of BXM?
The next chart shows the effect of monthly trading frictions ranging from 0.00% to 0.10% on the above terminal value of BXM. BXM beats SPY based on terminal value only for extremely low implementation friction.
Terminal value comparisons can be sensitive to start and stop dates. For another perspective, we look at the effect of trading friction on risk-adjusted monthly return.
The next chart shows the effect of monthly BXM trading frictions ranging from 0.00% to 0.20% on the ratio of average monthly return to standard deviation of monthly returns. BXM beats SPY based on this metric for monthly implementation frictions below about 0.15% of portfolio value.
How about the performance of PBP?
The final chart compares cumulative values of $1.00 initial investments in the gross BXM index as calculated by CBOE, SPY and PBP since the January 2008 option expiration date. Results illustrate the drag of implementation costs on the buy-write strategy, with BXM / SPY /PBP terminal values of $1.03 / $1.01 / $0.96.
The average monthly return for BXM / SPY /PBP over this period is 0.21% / 0.28% / 0.07%, with standard deviation of monthly returns 5.22% / 6.97% / 5.28%, indicating an average monthly PBP implementation friction of 0.14%. As indicated by the chart immediately above, this friction approximately balances any risk reduction offered by PBP compared to SPY.
In summary, evidence from simple tests suggests that trading frictions approximately offset any risk reduction benefits offered by systematically selling covered calls on a broad stock market index.
Cautions regarding findings include:
- As noted, the sample period for PBP is short.
- A covered call strategy executed with narrower indexes or individual stocks may perform differently.
- Investors utilizing small positions may experience trading frictions above the range considered.
You May Also Enjoy...
- Are Some Covered Calls More Profitable Than Others?
- Simple Tests of Sy Harding’s Seasonal Timing Strategy
It costs less than a single trading commission. Learn more here.
