Stock selection strategy

Simply is to buy a stock if:

1. The stock has traded for 36 months (IPOs pls excuse)
2. If todays price is less than the least price in the last 36 calendar months.
3. Price is less than 2/3rd of Book Value.
4. Debt/Equity less than one.

Stick with me while I take you through the logic behind the strategy. I will talk about my method of implementing it later. There are three different sources that led to this strategy.

1. Chandrakant Sampat
2. The stock: FCS Software
3. Dalal Street Small Cap 400 2007 (It has become 500 now. Back in 2007 it was 400).

The primary source for coming up with a metric like 36 months low is This article by Chandrakant Sampat (I am really grateful to him for his ideas).

Now, basically he asks you to buy stocks of good companies when they are at 8-10 year low. As I see it, there are some problems with this idea. If you follow this rule strictly (which I have absolutely no doubt in my mind would work really really well), you essentially need to sit out of the market for a very long time before such opportunities show up. You also cannot take advantage of periodic temporary (???) insanity that the market undergoes almost on a regular basis. Using the idea of Chandrakant Sampat as the starting point, I decided to crunch numbers and find out the level of multi year low (ie. is it 2 yr low or 7 year low etc.) that would be a good entry point for an investor like me (well, not quite. This was not for an investor like me, it was for me).

Now, I am not a very trigger happy guy (read trader) when it comes to buying stock but asking me to sit idle for 6-12 months twiddling my thumbs just doesn’t cut it. I found that 3 year low works best for me. It was the obvious and perfect choice for optimal tradeoff between frequency and value (again this was FOR ME). If I got very fidgety there were always some (mere;-) 24 month low stock to fall back upon. Hence the idea 36 months low.

Second source: The stock FCS Software. I think it was around July 2007 that I first got interested in this stock. At that time I was going through Warren Buffett letters to shareholders and tried to apply his ideas on the Indian stock market. This stock seemed to qualify all the requirements laid out by him: little or no debt, high return on equity at a very low price. When I ‘discovered’ FCS Software in July 2007, the price of the stock was around 90. Looking at its history I noticed that it was available at less than 30% three months back (in March 2007). It was quite clear to me that if I had ‘known’ about FCS Software, I would have definitely purchased it March 2007 . If I liked the stock at 90, I would have loved it at 65. This made it clear to me that I needed to do two things:

1. I needed to become ‘aware’ of more stocks (this was back when I was not aware of BSE/NSE Indices).
2. I need to find stocks on sale, when and while they are on sale, not three months later.

I went about trying to define a ‘deal’ and one of the obvious way of defining it was relative low (eg. 24/36 month low). If I go about finding stocks whose price TODAY is lower than any traded price in the last 36 months, the stock is available on deal TODAY. Also, if I keep looking at stocks that are available on deal today, I will also encounter (intentionally) some stocks I like (rather than discovering them by pure accident).

Third source: Dalal Street Small Cap 400 2007

I bought the Dalal Street Small Cap 400 2007 and was flipping through it trying to soak in stuff. It contained, among other stuff, the yearly lows and highs of a stock. I was amazed at a curious phenomenon - almost all the stocks had WILD swings and the difference between one years low and the next years high was mind boggling.

Let us take flip through that book randomly now. Let us take 15 random stocks from this list of 400 (if you are curious, the perl code used to generate this list is given at the end). The ranks of the stocks that we are interested are:

9,350,239,24,104,324,106,84,62,174,79,257,260,309,262

The following table illustrates the point I am trying to make (notice the difference between one year low and the next year high).

Rank	Company	2007 High/Low	2006 High/Low	2005 High/Low	2004 High/Low	2003 High/Low
9	Oudh Sugar Mills	91/53	295/80	175/69	140/17	27/6
350	Balmer Lawrie Van Leer	50/27	57/28	53/18	32/11	25/8
239	Vijay Textiles	8/4	9/3	385/6	361/29	27/16
24	Salora International	135/74	141/67	186/98	200/77	139/31
104	Sandesh	224/129	218/78	257/110	163/80	180/95
326	Ambalal Sarabhai Enterprises	25/12	16/9	21/9	16/5	13/3
106	Amar Remedies	64/29	102/42	68/41
84	Jayant Agro Organics	13/4	16/3	14/4	6/5	73/38
62	Fiem Industries	135/82	148/87
174	Super Sales India	347/167	375/160	229/87	125/21	47/9
79	Chemfab Alkalies	125/84	195/81	398/119	180/31	72/18
257	Hawkins Cookers	139/79	123/55	94/24	40/14	30/16
260	Gujarat Reclaim & Rubber Products	552/295	359/199	304/107	122/51	90/18
309	Simbhaoli Sugars	66/38	177/57	155/57	55/12	17/7
262	Jhagadia Copper	16/9	29/10	29/11	19/7	18/4

Here my logic runs something like this: Given that these stocks are so volatile, as long as I buy stocks on the lower half between 52 WK low and 52 Wk high (of last year, since at any time we will have data only upto the previous year) and wait patiently, the odds are with me that my investment will prove profitable. If the price is lower than last years low, the odds are further improved. Price at two years low is even better.

In short:

Multi Year Low Price +Diversification + Patience = Margin of Safety + Lots of Money

Now add to this Grahams simple rules for valuation and safety:

Multi Year Low Price +Diversification + 2/3rd Book Value + Low Debt + Patience = HUGE Margin of Safety + Lots of Money

I rest my case. I would love to hear what you think about this strategy. You can get in touch with me here.

Perl Code
————————————-
#!/usr/bin/perl
@smallcap = (1..400);
fisher_yates_shuffle (\@smallcap);
print join “,”, @smallcap[0..14];

sub fisher_yates_shuffle {
my $array = shift;
my $i;
for ($i = @$array; –$i; ) {
my $j = int rand ($i+1);
next if $i == $j;
@$array[$i,$j] = @$array[$j,$i];
}
}
--------------------------------------------------------
Written by Jaishankar · Filed Under Articles