Trading Around Option Expiration Days

Are there any stock market return/volatility anomalies around the equity option expiration (OE) date (third Friday of each month)? Potential anomalies include: (1) systematic differences in returns and volatilities before, on and after OE; and, (2) systematic differences in OE returns conditional on prior-month returns. To investigate, we examine close-to-close market returns from five trading days before to five trading days after OE. Using daily closing prices for the S&P 500 Index for January 1990 through May 2012 (268 OEs, with September 2001 excluded due to trading disruption) and for the iPath S&P 500 VIX Short-Term Futures ETN (VXX) during January 2009 through May 2012 (40 OEs), we find that:

The following chart shows average daily S&P 500 Index returns from five trading days before (OE-5) through five trading days after OE (OE+5) over the available sample period, with one standard deviation variability ranges. The mean daily return for all trading days in the sample is 0.03%. Results suggest a modest positive bias for a few days before and a negative bias just after OE. However, variability swamps average returns.

Is there a consistent pattern across subperiods?

The next chart summarizes average daily S&P 500 Index returns from OE-5 through OE+5 for three equal seven-year subperiods and since the beginning of 2011 relative to mean returns for respective subperiods (raw average daily return minus average return for all trading days in the subperiod). The number of observations for the first three subperiods is 83-84, with only 17 observations in the last subperiod. While results loosely support a narrative of relative strength before and relative weakness after OE, inconsistencies undermine belief in a reliable pattern.

Does OE return depend on market return since the prior OE?

The following scatter plot relates return on OE to return from the close on the preceding OE through the day before OE. For example, if two successive OE days are 5/15 and 6/19, the chart relates the return on 6/19 to the return during 5/15-6/18. The Pearson correlation between the two series is 0.02 and the R-squared statistic is 0.000, indicating no relationship. In other words, aggregate investor behavior is indifferent to prior-month return. Perhaps dynamic hedging obviates any OE day price pressure.

Does VXX exhibit a return pattern around OE?

The final chart shows average daily VXX returns from OE-5 through OE+5 over the available sample period, with one standard deviation variability ranges. The mean daily return for VXX over all trading days in the sample is -0.27%. Results suggest that VXX tends to be relatively strong just before OE and relatively weak just after OE. However, given the variability of returns, the available sample is small for inference (one or two new observations could alter results).

In summary, evidence from simple tests on available data offers slight support for belief in a modest positive (negative) bias for the broad stock market and near-term volatility futures a few days before (after) equity option expiration dates.

Cautions regarding findings include:

• The subsample for the S&P 500 Index since the beginning of 2011 is very short for useful inference.
• As noted, VXX returns exhibit strong variability, making the entire sample for VXX short for reliable inference.