Company Value

[slackjaw]

What is the best way to estimate the value of a company share? Can it be simply the value that gives an P/E of around 20? What else do people here take into account. Does anyone use a formula for this?

 

[fimmwolf]

I found this interesting bit of info on anther forum:

In The Intelligent Investor, Graham laid out an equation that was designed to help people value a growth company. The equation goes like this:

P = ProjEPS * (8.5 + (2*G)) * (4.4/AAA yield)

Looks scary, but it's not. Let me break it down for you.

P = the price of the company's stock. This is what the equation will tell you.

ProjEPS = the projected earnings per share. You're looking for next year's estimated earnings per share. (Get estimates here.)

8.5 Graham surmised that a zero growth stock should have a P/E multiple of 8.5. This reflects an average return of 12% per year. For what it's worth, I think this is a little shaky given the wide variety of circumstances leading to a company with zero growth. As you'll see, Graham put some qualifiers on this equation.

2*G where G is the long-term projected growth rate of a company's EPS. Graham said that you should be comfortable that the company will grow its earnings at this rate over the next seven to 10 years.

Unfortunately, most companies/analysts only give 5-year expected growth rates, so you'll have to use those.

4.4 This was Graham's benchmark for a required rate of return to invest, period. He surmised that at a minimum, an investor needed to be compensated for the effects of inflation and a small risk premium above that. You might be tempted to "play around" with the 4.4, but I keep it constant.


AAA yield This is the yield on the AAA corporate bonds. I like to use the 30-year composite yields, but they are difficult to find. Here is a link that's a few weeks dated, but it'll serve for now: http://www.bondresources.com/Corporate/Rates/AAA The 30-year yield on AAA corporates is about 6.25%.


Okay, so what does all this mean and why am I bothering you with it? Last week, I wrote about risk and expected returns. If you remember, we talked about how investing in one asset class (stocks vs. bonds vs. CDs, etc.) requires that you receive a greater rate of return relative to less risky classes. Graham's equation builds in an equity growth premium as well as an interest rate factor. It's not a magic formula that will solve all of your problems, but it does provide a decent data point to consider in our evaluation of a company's stock price.

Let's quickly do an example using Pfizer:
P = 1.59 * (8.5 + (2 * 19.5)) * (4.4/6.25)
P = 1.59 * (8.5 + (39)) * (.704)
P = 1.59 * 47.5 * .704
P = $53.16


According to this simple equation, Pfizer's value is about $53. The stock currently trades around $42.50. If you were to take this as gospel (and you better not), you might conclude that the stock is about 25% undervalued at current prices.


Why am I so cautious and advising you against using this as your "magic dust"? Well, it's only one model, it's imperfect, and it doesn't account for a host of other issues a company may deal with. It also doesn't address other kinds of valuations, such as discounted cash flow analysis, among others.


Graham provided four major caveats to his equation that I should share here.

In the book Small Stocks, Big Profits, Dr. Gerald Perritt describes these caveats as follows. In screening stocks, you should:

1. Eliminate all firms with negative earnings (losses).
2. Eliminate all firms with debt to total asset ratios greater than 0.60 (i.e., firms with total debt greater than 60% of total assets).
3. Eliminate all firms with share prices above net working capital per share.
4. Eliminate all firms with E/P (earnings divided by price) that are less than twice the AAA bond yield.

It's not the be all and end all, but it's a place to start.

Cheers

 

[dhukka]

What is the best way to estimate the value of a company share?

There are many ways. DCF's, Dividend Discount multiples, market multiples, ROE's but the best one of knwo of is contained in the book Market Wise written by Brian McNiven.

Can it be simply the value that gives an P/E of around 20?

No it can't, if you this method you will lose money, guaranteed.

What else do people here take into account. Does anyone use a formula for this?

Take a look at these threads here and here.

 

[michael_selway]

What is the best way to estimate the value of a company share? Can it be simply the value that gives an P/E of around 20? What else do people here take into account. Does anyone use a formula for this?

Hi first of all, why PE of 20? PE of 10? PE of 5?

What do you compare it against? Also Risk in investing in companies?

thx

MS

 

[fimmwolf]

There is also Ev/EBIDT

check it out

http://moneyterms.co.uk/ev_ebitda/?fr=y

 

[Ben10]

I found this interesting bit of info on anther forum:




It's not the be all and end all, but it's a place to start.

Cheers

Reading this book at the moment and recommend it

 

[johenmo]

Value as determined by what? The financial data? The market? Are you buying the company or shares in it?

I always thought the financial data is what made up value (that's bawhat I got taught, even at the overseas business school!). But in my short time I have seen companies with supposely good data plummet in share price.

Babcock and Brown had some "good data" but now I know that there is so much more to look at - who the directors are. where they get their money, the business model. I still have a few "useless" shares which I am holding as an experiment, albeit slightly costly.

It all depends on what you plan to do. Then you know what to look for to determine value.

 

[Lukas]

Hi slackjaw
I attempt to value ordinary shares using the dividend growth valuation models combined with Price/Earnings to firms fundamentals. Dividend models value the share by discounting expected future cash flows at the appropriate discount rate known as the required rate of return.

 

[Shrewd Crude]

slackjaw,
do alittle research on Net Present Value, NPV... its the best model...
In most of these models we have to make up some assumptions which we just cant get away from, ie we are dealing with revenue streams in the future...
fimmwolf's equation is probably a very good place to start because the equation is easy to use... but its not a totally practical equation because you are only using next years guidelines, what we want is next years guidelines, the year after, the year after that, and so on... the more info the better...
how can be base the value of a company on one years expected EPS? well we cant!... all these models are faulty in some way...