The end of April is imminent. Thus, also comes an end to the traditional good part of the year for equities. Sell in May is hitting the headlines again, reminding us of the relative sober period that is about to commence. Through time, investor and media attention for the Sell in May effect has increased. This poses an interesting question; 'Are investors anticipating the Sell in May effect?'
Sell in May Returns
Let's start with a quick overview of the Sell in May effect. The graph below shows the average returns for the S&P 500 index (NYSEARCA:SPY) during the winter period (November to April) and the summer period (May to October). In each of the four sub periods, the average return is higher during the winter period than during the summer period, highlighting the Sell in May Effect. The Sell in May effect is pretty outspoken, ranging from 3% to almost 6%. Also, the results show that the biggest chunk of the return on the S&P 500 index is realized during winter months. Selling your equities before going into the summer period would not have cost you all that much.
Over time the amount of attention for the Sell in May effect, this article being part of that development, seems to increase. And that poses an interesting question. Given the fact that it is pretty hard to miss the two pivotal points in the Sell and May effect, April and October, are investors anticipating the Sell in May effect over time? In other words has investor behavior changed?
In order to get an indication if behavior has changed, the focus of my piece lies on April and October. Hence, knowing that from May on the expected returns on the S&P 500 index are pretty marginal could have investors decide to sell (part of) their holdings already before the end of April. The same way, knowing that the expected equity return is pretty solid starting in November, could make investors anxious to get in a little earlier when it is still October. The effect of this change in behavior would then probably result in 1) a decrease over time of the average return in April and 2) a rise, again over time, in the average return in October.
Evolution over time
To find out if this is the case I calculated the average returns in April and October during a number of historical periods. The results are shown below. Let's look at April first. It is pretty obvious that the average return in April has not decreased in anticipation of May. Quite the opposite. The average return in April has increased aggressively. Actually, April was the best return month in the two most recent periods, from 1991 and 2002. For October, however, the results are mixed. There is no clear pattern in the average return over time. But that also means there is no real evidence that the behavior of investors has changed over time. Based on average calendar month returns, investors are not anticipating the Sell in May Effect.
Now there could be a caveat to this average monthly return approach. What if investors do anticipate the Sell in May effect, but only just before and after the event? In this case you would expect lower returns over time in the days surrounding the end of April and stronger returns over time in the days surrounding the start of November.
Turn of the month
To get a clue if this might be the case, I calculated the 'turn of the month return' for April and October for the same sub periods as above. The turn of the month return is the S&P 500 index return between the 25th of the any given month and the 6th of the next month. The results are shown in the graph below. They are quite interesting. In contrary to the calendar return analysis, there now are clear indications that investors are anticipating the Sell in May effect. For April, the average turn of the month return slowly decreases over time. Since 2002, the turn of the month return has turned negative, compared to a turn of the month return of 0.7% since 1928. Measured over all months, the turn of the month return tends to be positive. But, investors seem to have become more cautious towards the end of the winter period, surrounding the end of April and beginning of May. This could very well be in anticipation of the Sell in May pattern in stock markets.
The view that investors have changed their behavior in relation to the Sell in May effect becomes even more enticing when looking at the average turn of the month return for October. Over time the turn of the month return of October has increased quite significantly. So, surrounding the end of October, the end of the poor period for equities, investors already start buying, causing stock prices to rise.
Investor behavior has changed
To conclude; when looking at average calendar month returns there is no strong evidence that investors have changed their behavior in awareness of the Sell in May effect. Especially, the average return for April, which has increased over time, does not align with this hypothesis. But if we zoom in and take a look at the turn of the month returns of April and October we do see a shift over time. On average, the turn of the month return for April has come down and has even become negative in recent years. At the same time the average turn of the month return for October has increased quite aggressively. This is potential evidence, that once the attention for the Sell and May effect increases, investors do, in fact, anticipate the change in equity seasons surrounding the end of April and October.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.