Links | Sitemap | Search:

Home > Commentary > Archive > Market Commentary

Requiem for a May-November Romance

(May 4, 2006)

Dear Subscribers and Readers,

For those who had wanted to learn more about picking stocks and evaluating companies, it is that time of the month again – the time when we bring in our regular guest commentator Mr. Bill Rempel for a quick discussion of his methods and thought processes. In this commentary, however, Bill will take a different route – and try to put his own mark on the “Sell in May” strategy that many investors are so fond of. Read it – you'll be surprised. For those who were looking for an analysis of an individual stock or industry, I have “invited” him back again for another mid-week commentary next week, since this author has an actuarial exam next Friday (which unfortunately means my upcoming weekend commentary will be a little bit abbreviated as well).

By the way, please continue to cast your votes on the prospects of Dell. Is Dell a buy, a hold, or a sell? So far, the sellers are clearly getting the upper-hand!

Without further ado, following is biography of Bill:

Bill Rempel (aka nodoodahs) is an active poster on the MarketThoughts forum as well as a few others around the web. Bill is a regular, monthly guest commentator on our website (see “Coca-Cola Still a Sell” for his last individual stock commentary). Bill graduated from Caddo Magnet High School (a high school for nerds) in back in 1985 and proceeded to learn the hard way when he drank his way out of a scholarship to Tulane later that year. After a few years of sweating for a living, he decided to go back to school, and graduated from LSU-Shreveport in 1995 with a Bachelors in Mathematics - all the while working the overnight shift stocking shelves in a grocery store.

Post-college, Bill has been in the P&C insurance industry as an actuary, product manager, and pricing manager. Bill and his wife Millie are amateur investors with a variety of holdings, but they prefer to buy and hold value investments. In typical "value" style, they live cheap, driving old cars and preferring to save or invest instead of buying fancy "stuff."

Disclaimer: This commentary is solely meant for education purposes and is not intended as investment advice. Please note that the opinions expressed in this commentary are those of the individual author and do not necessarily represent the opinion of MarketThoughts LLC or its management.

Once again, it's time for another round of "Sell in May and go away." Of course, we're still about half a year away from "Sell on Rosh Hashanah, Buy on Yom Kippur," but I digress. Or do I? Never one to leave well enough alone, despite the wealth of writing on this subject, I thought I'd see if a deeper look could be of some interest.

Assuming a totally mechanical approach, where one bought the Dow Jones Industrial Average (DJI from here on out) on the first day of trading in November, and sold on the first day of May the following year, the first issue one comes across is what to do with the cash for six months. Good question! The possibilities include cash, fixed income like the long bond or equivalent, etcetera. For the purpose of this analysis, I am assuming a 0% return for the time period that one is "out of market." To consider anything else introduces the seasonality component of another investment, which is beyond the ken of this article – I want to focus specifically on market timing of the DJI.

A brief reminder of what I am not considering in this examination: dividends and tax issues. I'm also not going to rehash the history of the saying, since that's been done.

How does "Sell in May" do?

Since 1950, "Sell in May" has delivered a 53.58-fold return on your money, or an average increase of 7.37% annualized. This compares to the fully invested buy-and-hold gain of 53.04-fold or 7.35% annualized.

The first obvious point is that the majority of historical gains take place in winter, and that summer trading tends to be flat. The second obvious point is that your return for the annual "out of market" experience doesn't have to be too great in order to beat your tax and dividend losses.

Since 1980, "Sell in May" has delivered a 9.48-fold return on your money, or an average increase of 9.04% annualized. This compares to the fully invested buy-and-hold gain of 13.91-fold or 10.66% annualized.

Now we see that for recent periods, the "Sell in May" hasn't worked quite as well as it used to. Given that explanation numero uno is always "everyone's on vacation," it stands to reason that as markets are more connected electronically, this pattern would weaken. However, the first obvious point remains: summer trading (broadly defined as May 1 through November 1) is flatter than winter trading (broadly defined as November 1 through May 1).

But I don't like broad definitions!

Narrowing the Concept

If the trade for a six-month period is, on average, worse than for the remainder, then how does that break down into constituent parts? Note for the following table, the return for a given month is the change from opening to opening, i.e. May's return is from the open of the first day of May to the open of the first day of June.*

Chart of Monthly DJI Returns from 1950-Present

Month	Lowest Return	Median Return	Average Return	Highest Return	Standard Deviation of Returns	Percent of Positive Returns
Jan	-8.4%	1.1%	1.3%	14.4%	4.9%	68.4%
Feb	-7.4%	0.8%	0.2%	8.8%	3.3%	57.9%
Mar	-9.0%	1.4%	0.9%	7.3%	3.2%	63.2%
Apr	-6.3%	1.4%	1.8%	10.7%	3.8%	61.4%
May	-7.8%	0.3%	0.1%	8.3%	3.4%	51.8%
Jun	-8.5%	0.0%	-0.1%	6.2%	3.3%	50.0%
Jul	-6.6%	0.9%	1.0%	9.0%	3.9%	60.7%
Aug	-14.5%	0.6%	-0.1%	11.5%	4.8%	55.4%
Sep	-12.3%	-1.1%	-1.1%	7.3%	4.2%	35.7%
Oct	-23.2%	0.5%	0.6%	11.6%	5.2%	57.1%
Nov	-14.2%	2.2%	1.7%	10.1%	4.6%	66.1%
Dec	-6.3%	1.8%	1.8%	9.5%	3.0%	73.2%

I've highlighted some areas of interest.

October is known for its crashes, and therein we have the lowest "lowest return" number and the highest standard deviation of returns, indicating the great variability of October.

The boxed section is the "Sell in May" portion, and both median and average returns suggest that these six months are definitely the bottom half of the year.

But wow! Look at September! From 1950 to the present, this is the only month that has negative median and average returns! Also, it is the only month that has less than a 50/50 chance of making a gain!

Chart of Monthly DJI Returns from 1980-Present

Month	Lowest Return	Median Return	Average Return	Highest Return	Standard Deviation of Returns	Percent of Positive Returns
Jan	-5.9%	1.0%	1.6%	13.7%	4.3%	70.4%
Feb	-7.4%	1.4%	0.8%	8.8%	3.9%	63.0%
Mar	-9.0%	1.3%	0.5%	7.3%	3.9%	63.0%
Apr	-4.4%	1.3%	1.9%	9.8%	3.8%	55.6%
May	-5.5%	1.5%	1.4%	8.3%	3.2%	61.5%
Jun	-6.9%	0.5%	0.3%	5.5%	3.0%	57.7%
Jul	-6.2%	0.7%	1.0%	9.0%	3.9%	57.7%
Aug	-14.5%	0.5%	-0.1%	11.5%	5.9%	53.8%
Sep	-12.3%	-1.1%	-1.6%	4.7%	4.3%	30.8%
Oct	-23.2%	1.7%	1.2%	11.6%	6.5%	61.5%
Nov	-8.0%	3.7%	2.3%	8.4%	4.5%	73.1%
Dec	-6.3%	1.8%	1.7%	9.5%	3.3%	69.2%

Checking out the 1980-present chart, most of the same commentary holds. Note however that the returns for October and May are starting to appear better, relative to the rest of the year.

Again, the problem isn't the summer, the problem is September!

Returns by Month and Year

		Average Return by Year
Month	1980	1981	1982	1983	1984	1985	1986	1987	1988	1989
Jan	4.4%	-1.8%	-1.0%	2.8%	-3.0%	6.2%	1.6%	13.7%	0.4%	8.0%
Feb	-1.5%	2.9%	-4.8%	3.4%	-5.4%	-0.2%	8.8%	3.1%	5.8%	-3.6%
Mar	-9.0%	3.0%	-0.2%	1.6%	0.9%	-1.3%	6.4%	3.6%	-4.0%	1.6%
Apr	4.0%	-0.6%	3.1%	8.4%	0.5%	-0.7%	-1.9%	-0.8%	2.2%	5.5%
May	4.1%	-0.6%	-3.4%	-2.1%	-5.5%	4.6%	5.2%	0.2%	-0.1%	2.5%
Jun	2.0%	-1.5%	-0.9%	1.8%	2.4%	1.5%	0.9%	5.5%	5.4%	-1.6%
Jul	7.8%	-2.5%	-0.4%	-1.9%	-1.5%	0.9%	-6.2%	6.3%	-0.6%	9.0%
Aug	-0.3%	-7.4%	11.5%	1.4%	9.6%	-1.0%	6.9%	3.5%	-4.8%	2.9%
Sep	0.0%	-3.6%	-0.6%	1.4%	-1.3%	-0.4%	-6.9%	-2.5%	4.2%	-1.6%
Oct	-0.9%	0.7%	10.7%	-0.6%	0.1%	3.4%	6.2%	-23.2%	1.7%	-1.8%
Nov	7.2%	3.9%	4.8%	4.1%	-1.5%	7.1%	1.9%	-8.0%	-1.6%	2.3%
Dec	-2.7%	-1.6%	0.7%	-1.4%	1.9%	5.1%	-0.9%	6.4%	2.5%	1.7%

		Average Return by Year
Month	1990	1991	1992	1993	1994	1995	1996	1997	1998	1999
Jan	-5.9%	3.9%	1.7%	0.3%	6.0%	0.2%	5.4%	5.7%	1.0%	2.1%
Feb	1.4%	5.3%	1.4%	1.8%	-3.7%	4.3%	1.7%	0.9%	6.8%	-1.0%
Mar	3.0%	1.1%	-1.0%	1.9%	-5.2%	3.7%	1.9%	-4.3%	3.4%	5.5%
Apr	-1.9%	-0.9%	3.8%	-0.2%	1.3%	3.9%	-0.3%	6.5%	3.3%	9.8%
May	8.3%	4.8%	1.1%	2.9%	2.1%	3.3%	1.3%	4.6%	-2.2%	-2.2%
Jun	0.1%	-3.8%	-2.3%	-0.3%	-3.5%	2.0%	0.2%	4.7%	1.2%	4.0%
Jul	0.9%	3.9%	2.3%	0.7%	3.8%	3.3%	-2.2%	7.2%	-1.6%	-2.9%
Aug	-10.0%	0.6%	-4.0%	3.2%	4.0%	-2.1%	1.6%	-7.0%	-14.5%	1.6%
Sep	-6.2%	-0.9%	0.4%	-2.6%	-1.8%	3.9%	4.7%	3.8%	2.2%	-4.6%
Oct	-0.4%	1.7%	-1.4%	3.5%	1.7%	-0.7%	2.5%	-6.3%	11.6%	3.8%
Nov	4.8%	-5.7%	2.4%	0.1%	-4.3%	6.7%	8.2%	5.1%	4.6%	1.4%
Dec	2.9%	9.5%	-0.1%	1.9%	2.5%	0.8%	-1.1%	1.1%	1.9%	5.7%

		Average Return by Year
Month	2000	2001	2002	2003	2004	2005	2006	2007	2008	2009
Jan	-4.9%	0.9%	-1.0%	-3.5%	0.3%	-2.7%	1.3%
Feb	-7.4%	-3.6%	1.9%	-2.0%	0.9%	2.9%	1.2%
Mar	7.3%	-5.9%	2.9%	1.3%	-2.1%	-2.7%	1.1%
Apr	-1.0%	8.7%	-4.4%	6.1%	-1.3%	-3.0%	2.3%
May	-2.0%	1.7%	-0.2%	4.4%	-0.4%	2.7%
Jun	-0.8%	-3.7%	-6.9%	1.5%	2.4%	-1.8%
Jul	0.7%	0.2%	-5.5%	2.8%	-2.8%	3.6%
Aug	6.6%	-5.5%	-0.8%	2.0%	0.3%	-1.5%
Sep	-5.0%	-11.1%	-12.3%	-1.5%	-0.9%	0.8%
Oct	2.9%	2.7%	10.6%	5.7%	-0.5%	-1.2%
Nov	-5.0%	8.4%	6.0%	-0.2%	4.0%	3.5%
Dec	3.6%	1.8%	-6.3%	6.8%	3.4%	-0.8%

"I won't see you in September"

In honor of finding this anomaly, I propose we have a new name and a new strategy: "I won't see you in September" and the 11-month market timing strategy.

How does it do?

Chart of 1950-Present Comparative Returns for Different DJI Strategies

	Buy/Hold	Sell in May	I won't see you in September
Increase	53.04	53.58	101.80
Annualized	7.35%	7.37%	8.61%

Chart of 1980-Present Comparative Returns for Different DJI Strategies

	Buy/Hold	Sell in May	I won't see you in September
Increase	13.91	9.48	21.79
Annualized	10.66%	9.04%	12.58%

This is a much better strategy. On average since 1950 or since 1980, "I won't see you in September" beats the pants off of "Sell in May."

SO exactly what is going on in September? Well, Rosh Hashanah doesn't always fall in September, even though it always falls on the same days of the year (in the Hebrew calendar, that is), and there's not a full month between it and Yom Kippur, so it doesn't make for a good observance of this trend, not to mention the fact that Yahoo! doesn't do a conversion for the dates. I could have called it "Sell on Army Day in Chile and buy it back on Armed Forces Day in South Korea," but that just doesn't ring in my ears. Hence, "I won't see you in September."

* The table at forum/viewtopic.php?t=384 was off by one month, that is, the December return should have been on January, etcetera. My apologies.

Article Tools