With arrival of October, there were bound to be some noise among investors and traders about the October Effect on stock markets. I am adding to the noise, with this data-driven post. :-)
If you haven’t heard about this effect before, then here is a short primer borrowed from Investopedia:
October Effect is a theory that stocks tend to decline during the month of October. It is considered mainly to be a psychological expectation rather than an actual phenomenon. Most statistics go against the theory. Some investors may be nervous during October because the dates of some large historical market crashes (in US) occurred during this month.
In US, markets dropped 24% in two days in October 1929 – still the worst 2-day drop in American history. Then again in October 1987, the Dow dropped a record 22% in just one day. Come October 2008, and Dow was down over 22% in eight trading days.
Not surprisingly, there is a quote about this effect too – given by the famous Mark Twain:
“October: This is one of the peculiarly dangerous months to speculate in stocks. The others are July, January, September, April, November, May, March, June, December, August and February.”
It is because of these crashes in historically well-documented periods (like recessions and depressions), that people tend to cling on to these magnetic phrases like ‘October-Effect’, ‘Black-Monday’, etc. and become nervous when such months or days are around the corner. Humans also have this psychological tendency to pick up random events and generalize them. This also adds to the popularity of these terms.
But enough of foreign data. Lets see how October has fared for Indian markets.
Since 2000, the monthly returns of Sensex have been as follows:
Now I can’t find any ‘October Effect’ here. On the contrary, I can see a March effect. ;-)
Lets go deeper into the past – upto Jan 1990:
Here I can see something, which does seem like an effect of sorts – atleast when we look at the averages. But look at the 3rd column and you will realize that like October, even March has had similar number of negative-return months (16 out of 26). Even January and April have been close (with 13 out of 26). The impact of any such ‘October Effect’ vaporizes.
But now, lets think about it logically. When you group market returns according to months, it’s obvious that some months will come on top and others at bottom – that is how mathematics works. Isn’t it? Just trying to find a trend for the heck of it is a good academic exercise*.
But it is an altogether different matter to use this or any other trend, to make money in stock markets on a regular basis.
Successful traders manage doing it on the basis of short-term trends.
Successful investors manage doing it, on the basis of long-termtrends.
As for the 'Rest' of the people… they are categorised as ‘rest of the...’ for a reason. ;-)
* - I love doing this.
In a smaller and ideal(er) world, something like ‘October Effect’ will be a reality. It is like a self-reinforcing idea. So if everyone thinks that it is true, then everyone will act accordingly. Eventually, everybody’s actions will lead to the result, which was expected at the first place (in this case: October Effect).
It is similar to the basic concept that is taught about inflation in theoretical economics class:
A key factor in determining inflation is people’s expectations of future inflation.
So if firms and consumers expect future inflation then it can become a self-fulfilling prophecy. If workers expect future inflation, they are more likely to bargain for higher wages to compensate for the increased cost of living. If workers can successfully bargain for higher wages, this will contribute towards inflation.
Now apart from the term 'October Effect', there is another interesting market aphorism:
"Sell in May and go away.”
Its more popular in US than in India but even then, many people do give it some weightage. The idea, of course, is that stocks tend to underperform during the six-month period beginning in May and ending October. So if it were an ideal world and the above aphorism remained valid, then it will be pretty similar to the buy low and sell high philosophy. It suggests that investors sell of their portfolios at the start of May and come back and buy again in October-end.
Sounds interesting? Its like selling expensive stocks in May; Then going away on a 6 month long vacation and coming back to a market, which has fallen and has good stocks available at cheap prices. Wouldn’t that be lovely?
I wish it were true. :-)
Whether it is or not is something that we will try answering a little later.
I did some online searches and found that the phrase - 'Sell in May and go away' is based on an old stock market adage - "Sell in May and go away, come back on St. Leger's Day.”
According to online references, this proverb can be traced back to the old era in Great Britain, when traders closed their positions to enjoy the summer season and returned only after the St Leger's - supposedly, the last race of the season, held in September. This has lead many people to believe that the period between May and October is not good for investing!! Yes. It might sound hilarious, but there are people who did believe it.
And you will be surprised to know that there have been studies to prove it. A famous study published in the American Economic Review in 2002 found that, this phenomenon does exist and that returns on stock markets in 36 out of 37 countries studied from 1970 to 1998 were higher in the November to April period than they were in the May to October period. But then there was another study published in 2004, which attributed the higher return to a couple of extreme data points of October 1987 crash the August 1998 collapse of Long Term Capital Markets. These and few other studies found no exploitable opportunity in US markets based on the. However, what it did find was that coincidently, many major economic and political events seem to take place during May to October period – which result in extreme market movements.
So coming back to India, does the modified adage “Sell in May. Buy in October (November)” work here?
The average return of the broader market - Sensex - between May and October since 2000 has been 9.35% compared with 6.92% delivered between November and April during the same period. But then, there are other periods like July – December and August – January, which have given average returns upto 13%.
Apart from that, the number of positive and negative returns periods are almost identical (10-11 for Positive and 5 for Negative).
Interestingly, the falls have been greater during the May-October period. As can be seen below, the list is dominated by May-October period and not by November-April period.
Now in another attempt to see whether Sell in May and Buy in November really does work in a practical scenario or not, lets consider 2 investors who wanted to invest Rs 10,000 in in stock markets in 1995.
Investor A is a typical buy-and-hold investor and does not believe in supposedly insane statements like Sell in May and Buy in November. He just invests Rs 10,000 in Sensex on 1st November 1995 and forgets about it.
Investor B is totally convinced about the May-November philosophy and holds his investments between November 1 and April 30. He then sells everything on May 1st and holds cash for the remaining 6 months before buying again on November 1st.
The figure below shows the results of the two investor's investments upto September 2015.
A buy-and-hold strategy of Investor A, would have turned an Rs 10,000 investment into Rs 74,975.
In comparison, Investor B who sold in May and bought back in November would have ended up with Rs 40,000. And this is when I have ignored costs, taxes related to transactions and also, the efforts involved in trading frequently. Add to this the fact that investor B would have also lost some amount of dividends, which would be available to Investor A for staying invested, we know what is the conclusion here. Simplicity works in your get-rich endeavors.
But one might argue that the bull market of 2003-2008 might have distorted the results in favor of buy and hold investor. But then, the crash of 2008-2009 was amongst the worst ever and gave sufficient variety to the data sampling. Then we also covered the dot-com crash of 2000s too. But nobody can prove or disprove buy and hold’s superiority over May-October Investing. What I did was only for a scenario, which started in 1995. Its possible that in another scenario, which might start in say 2002, the May-November logic might give better returns.
But what is clear from this particular scenario is that the buy-and-hold strategy is considerably simpler, doesn’t sacrifice dividends and significantly eliminates the transactional costs. But if you are asking me to pass a judgment in favor of either of these two approaches, then that’s not going to happen. And that is because, even though we have some data to prove that theories like ‘October Effect’ and ‘Selling in May and Buying in November’ are not worth much, there is no conclusive evidence that it holds true at all times.
For common investors, it’s best to stick with simpler things and ignoring complex and nice-sounding theories. In most cases, these might not work as a common investor does not have the time or resource to implement these strategies.