How to value stocks series
The second method I use to value a stock is by using Benjamin Graham’s formula from The Intelligent Investor.
With the extremely popular free Ben Graham stock spreadsheet I offer, the stock valuation method deserves a closer look.
Benjamin Graham Formula
The original formula from Security Analysis is
where V is the intrinsic value, EPS is the trailing 12 month EPS, 8.5 is the PE ratio of a stock with 0% growth and g being the growth rate for the next 7-10 years.
However, this formula was later revised as Graham included a required rate of return.
The formula is essentially the same except the number 4.4 is what Graham determined to be his minimum required rate of return. At the time of around 1962 when Graham was publicizing his works, the risk free interest rate was 4.4% but to adjust to the present, we divide this number by today’s AAA corporate bond rate, represented by Y in the formula above.
(credit to wikipedia for the formula images)
Adjust Earnings Per Share
But intrinsic value shouldn’t be calculated based on a single 12 month period which is why I have the EPS automatically adjusted to a normalized number ignoring one time huge or depressed earnings based on 5 year or 10 year history depending on the company you are looking at.
EPS is never really a good number on its own as it is highly prone to manipulation with modern accounting methods. Another reason why you have to always normalize EPS is because management will never understate earnings on purpose. While companies may follow accounting procedures which inflates earnings, they will never go out of their way to make it lower than it is.
Another variation of the formula will use the projected EPS but unless it is a pure growth stock with exponential growth like characteristics, the stock value will become absurdly high.
EPS by analysts are also always over optimistic, so by following Wall Street guidance, you’re starting off on the wrong foot.
Adjust Growth Rate
The drawback of the Benjamin Graham formula is that growth is a big element of the overall valuation.
You can change 8.5 to whatever you feel is the correct PE for a no growth company. Depending on your conservativeness, anything between 7 and 8.5 should be fine.
For the actual growth rate, if convenience is important, you could just use the analyst 5yr predictions from Yahoo or other sites, but for most value stocks that I search for, predictability is important so a regression of the historical EPS to project the following year is a method I like to use.
The “2 x G” however, is quite aggressive. So I’ve recently reduced the multiplier to 1.5 instead of 2.
Corporate Bond Rate
I currently have the stock value calculator set up to use the 20 year A corporate rate which is just above 6%. This provides a slightly more conservative intrinsic value than the 20 year AAA or AA.
Final Adjusted Benjamin Graham Formula
So by making the adjustments, the new formula is now
Testing the Formula
Testing this equation on Microsoft, the inputs are
- Normalized EPS = $1.40
- g = 12.6%
- Y = 6.05%
which results in an Ben Graham intrinsic value of $29.10. Current price as of writing is $29.41.
Results look pretty good, but not all companies are as predictable or stable as MSFT so the stock valuation could be a coincidence.
A growing company DLB.
- Normalized EPS = $2.1
- g = 17%
- Y = 6.05%
- Intrinsic value = $53.56
- Current price = $44.72
What about growth story AAPL?
- Normalized EPS = $7.6
- g = 18.6%
- Y = 6.05%
- Intrinsic value = $207.41 (much different to my dcf valuation of AAPL)
- Current price = $199.91
The results aren’t all too bad but obviously, you will need to be careful of your inputs. And never forget margin of safety.
Disclosure
No positions at time of writing.
December 1st, 2009 at 10:37 pm
Hi Jae, instead of using normalized EPS what about owner earnings per share?
December 1st, 2009 at 10:57 pm
Hey Jae, no matter how I plug in the numbers in the examples I cannot get the correct intrinsic value from the examples.
For the MSFT valuation it should be V = [1.4 (7 + 1.5(12.6)) x 4.4]/6.05
Please correct me if I am mistaken!
December 1st, 2009 at 11:12 pm
@ Ryan
I never thought about trying it with owner earnings. I don’t see why it won’t work. Use the normalized owner earnings and use the owner earnings growth rate and plug it in the formula. I’m gonna try it right now and see what I get.
December 1st, 2009 at 11:34 pm
@ Andrew
Keep in mind that the formula looks real simple but you need to expand it.
If I expand the numerator it becomes
4.4*EPS*8.5 + 4.4*EPS*2g
then divide the whole thing by 6.05.
In excel the formula is
=1.4*(7+(1.5*12.6)*(4.4/6.05))
December 2nd, 2009 at 8:01 pm
why did my last comment get deleted?
December 2nd, 2009 at 8:09 pm
Jae, I plugged in the numbers to both of your equations and results are the same as what I have calculated, 26.37 and not 29.10.
December 2nd, 2009 at 10:51 pm
Andrew,
I think you’re making a mistake with the order of operations somewhere.
Just enter “=1.4*(7+(1.5*12.6)*(4.4/6.05))” into excel and you will get the number.
I didn’t delete any comments by the way.
December 4th, 2009 at 12:52 am
Hi Jae, thanks for the clarification, I got the same number using 1.4*(7+(1.5*12.6)*(4.4/6.05)). Just wanted to point out V= [EPS*(7*1.5g)*4.4]/Y (the equation you have on top) is not the same as the formula you provided in the excel!
Anyways really enjoy reading your blog keep up the good work!
December 4th, 2009 at 1:40 am
Andrew,
You’re right!
I made the mistake and never re-checked it after all this time.
Thanks for pointing it out. I’ll fix it and send it out in the next update.
December 4th, 2009 at 10:49 am
hi jae, although i feel okay with your adjustments to the formula. but then intrinsically, you’re changing benjamin graham’s formula. as we all know, buffett is the most disciplined follower of graham. if you take a look at his recent BNI acquisition. buffett still uses graham’s formula to place a value on BNI. we get a value around $93 and buffett paid $100.
when you adjust your formula. we get a value for BNI around $65. then would you be suggesting that buffett overpaid for BNI (which is not his usual practice)?
December 4th, 2009 at 10:55 am
David,
Based on my assumptions, Buffett wouldn’t use Graham’s formula as it depends heavily on EPS which is something Buffett has clearly expressed is not correct. Even if he decided to use his Owner earnings as the EPS, that still changed the formula.
The formula is only guide. There is no right and wrong way, but for my investing, the standard Graham formula is not correct.
BNI is a very different story and is much deeper than a valuation. Buffett bought it as an inflation protection investment as well as its hidden assets which is not shown on the balance sheet.
So on paper, I think he overpaid, but in reality, he probably got a good deal. Also remember his See’s Candies purchase was not a cheap one at the time.
December 18th, 2009 at 3:49 pm
Hi jae,
just wanted to know if Grahams formula can be used for UK stocks?
Great site by the way.
December 19th, 2009 at 2:59 pm
Luke,
Yes the formula works for all companies
March 4th, 2010 at 2:41 pm
Good post. I have 3 questions.
1) Does “The Intelligent Investor” explain the methodology behind the variables in the formula, specifically why was 2 used in 2g (ie – I understand 2 is modifying the PE ratio to reflect the growth rate, but what makes 2 the magic number to adjust the growth rate? or 1.5 for that matter)
2) Why are we multiplying the percentages by 100 (ie 12% is 12 in your formula and not .12)?
3) Can you explain the original equation is multiplied by 4.4%/y. I see your explanation above, but don’t follow the mathematical relationship.
If all of this is covered in Grahm’s book, then just tell me to defer to the book.
Thanks,
-Rojo
March 12th, 2010 at 6:59 am
Actually Buffett ended up paying about $93/share for BNI if you include the 22% he bought on the open market before the buyout offer. Those shares were purchased at a cost basis of something like $75.
May 25th, 2010 at 6:51 pm
Who are you?
May 25th, 2010 at 6:52 pm
…Andrew, that is.
June 22nd, 2010 at 9:26 am
I’d like to point out that the calculations on this page are incorrect.
So instead of:
=1,4*(7+(1,5*12,6)*(4,4/6,05))
it should be:
=1,4*(7+(1,5*12,6))*(4,4/6,05)
See http://www.fool.com/portfolios/rulemaker/2001/rulemaker011031.htm for details.
June 26th, 2010 at 7:51 am
Dear Jae,
At the outset, I would thank you to provide us with such a valuable source of investment information. I am hugely fascinated by your excel spreadsheet intrinsic value calculator. However, my problem is that I am in India and I do not know a database from where I could pull up the data nto your spreadsheet (after i buy it). Could you please somehow use your research and help me with this problem. Secondly, I have personally developed a Graham formula and DCF valuation worksheet, where my biggest problem is the calculation of growth rate. My question is that whether five years growth data of free cash flow would be enough to arrive at a meaningful value of the stock. Any more suggestions would be helpful. Thanks.
June 26th, 2010 at 8:34 pm
@ Imran,
I get asked a lot about whether the spreadsheets will work outside of USA and the simple answer is no.
As far as I know, there is no other country in the world that offers the level of transparency required by public companies as well as the vast amount of resources available by 3rd party sites.
Regarding the growth rate calculation, it is very subjective but you wouldn’t want to use the FCF growth rate for the graham calculation because the Graham formula is based off earnings. You would have to use the EPS growth for that.