Finally, last week, I thought of coming up with some systematic formula for better and more effective valuation.
Some reading as well as ideas given by forum members, particularly @Antonym were helpful in doing this fast.
Why I am sharing this is;
First, to give out something to the forum as I am also learning a lot from it,
Second, to get views of other so that I can improve on this,
Third; when I try to explain something to others, I understand it better.
Since my approach to this is systematic, I will try to explain it step by step.
Here we go.....
How much a share worth?
Traditionally, the easiest and quickest way I found is to look at one figure in the financial reports of the entity (I mean the company or the group);
the book value or Net Asset Value per Share (I call it NAV).
So my first version of formula is;
Value = NAV
But we cannot fully depend on this number due to few limitations; such as
* NAV shows only the result of past activities and it does not fully represent what is going on and what will happen next
* NAV itself may not be reliable sometimes due to inconsistent practices in accounting (this is my view; arguable), etc
Therefore, I will bring in another factor; how much profit the business generates annually; Earnings per share (EPS).
But there is a problem;
How many years of EPS are we going to consider?
This depends on how we think of the business environment. Here, for the time being I assume the current status can continue for another 5 years.
So, my second formula for,
5 year projection is;
Value = NAV + 5*EPS
But there is an issue. A rupee worth today will not worth a rupee in 5 years time. So we have to use a discount rate. I take the yield rate of the next best investment options as the discount rate;
Average FD rate (Fixed deposit) which is 12% (Pls let me know if you think of a different rate)
So a simple calculation of NPV (Net Present Value) gives us the third formula for,
5 year projection
Value = 0.6*NAV + 3*EPS
Well there is another thing investors consider; Dividends.
In fact, there is a saying 'a bird in the hand worth two in the tree' (Something like that).
Actually, it seems, this is one of the most important factors the industry consider.
Let's say the share get an annual dividend of DS.
So, with NPV (Net Preset Value), our next formula for,
5 year projection is;
Value = 0.6*NAV + 3*(EPS-DS) + 4*DS
(; since DS is paid out of EPS, now EPS has become EPS-DS)
Now, there is another situation;
due to many factors, EPS and DS are not constants, they fluctuate and they undergo trends.
In order to accommodate the earning and dividend trend, I would use trend adjusted average EPS and DS (taEPS, taDS)
So our formula for;
5 year projection becomes;
Value = 0.6*NAV + 3*(taEPS-taDS) + 4*taDS
In order to accommodate fluctuations I consider 'standard deviation' of EPS and DS (sd); this lead us to two values Lower value and Higher value (high Value, low Value);
So our formulas for;
5 year projection are;
high Value = 0.6*NAV + 3*(taEPSpsd-taDSpsd) + 4*taDSpsd
low Value = 0.6*NAV + 3*(taEPSmsd-taDSmsd) + 4*taDSmsd
(psd mean plus sd (sd added), msd means minus sd).
So those are my (for the time being) final formulas for 5 year projected value.
But I am a little uncomfortable assigning a higher weight to NPV.
Furthermore, both NAV and EPS are just numbers only so they are not indisputable; the only tangible and indisputable entity here is DS; so I prefer it getting a higher weight.
Therefore, I consider other time frame projections and arrive at the formulas for;
10 year projection;
high Value = 0.3*NAV + 3*(taEPSpsd-taDSpsd) + 6*taDSpsd
low Value = 0.3*NAV + 3*(taEPSmsd-taDSmsd) + 6*taDSmsd
These are my selected formulas for the time being;
By applying to some sample shares, they showed some meaningful results.
Will come up few actual results later......(Still trying to find EPS and DS from quarterly reports; if somebody already have them (numbers copied in a spread sheet) pls share with me)
Pls let me know if you feel if this approach look sensible.