Throughout 1988 and 1989, Warren Buffett acquired more than $1 billion of Coca-Cola (KO) stock. At the time, Wall Street thought he was downright crazy. After all, Wall Street scrutinized the purchase and deduced that Buffett has paid way too much for earnings and the stock price was high—having run up 18% a year for eight years.
In 1988, Wall Street said Coca-Cola was a bad stock to buy. Warren Buffett thought it was a wonderful business to own. The results speak for themselves; so, let's look at the reasoning behind Warren Buffett's most famous purchase.
You can follow along by downloading the PDF version of the analysis here. (Yes, it's free.)
From 1978 through 1987, Coca-Cola's Free Cash Flow grew at a median rate of 21.8% a year. Buffett himself says we should not take yearly results too seriously, so we focus on multi-year results. Then again, Coca-Cola's Free Cash Flow grew fairly steadily each year—a definite plus!
Coca-Cola's Shareholder Equity had been growing about 7.8% a year. Not stellar by any means, but it was consistent and predictable—both staples in the Buffett approach to investing. The growth rate of Shareholder Equity becomes critically important only when you expect your company to close up shop in the next twenty years—clearly not in the stars for Coca-Cola.
Coca-Cola had a median CROIC of 9.3% for ten years. For every dollar of capital invested in the company, Coca-Cola was generating $0.09 of cash. As I mentioned in this discussion of CROIC, I prefer to see CROIC above 13%. Any lower and the numbers become fragile. Then again, Coca-Cola was a special situation because of its brand and moat. In 1988, Coca-Cola was anything but fragile.
A company that needs no introduction, Coca-Cola was the company in the beverage industry...and in the world. It dominated the market and had no serious competition. Picture a world where there was practically no Pepsi, Snapple, or bottled water on the shelves—just Coke. That is pretty much 1988.
In 1988, you would have been hard pressed to find a more well-known name than Coca-Cola. Now that is moat.
Assuming the company could continue to grow Free Cash Flow at 21.8% a year for ten years, and then slowed to 5% thereafter, and assuming Buffett wanted a 15% or more average annual return, you could value Coca-Cola at $22.3 billion, or $59.16 a share in 1988.
For new readers: The $22.3 billion is made up of $2.09 billion of Shareholder Equity and the net present value of the estimated $98.89 billion of future cash flow, discounted at 15% for a handsome return. Here is why we look at these numbers. See Calculating The Value Of A Business for a more detailed explanation of the calculation.
Of course, you shouldn't pay full price for a company—even one as solid as Coca-Cola. If the future is a little less rosy than you projected, your returns head south. So, you need a discount. Being an industry leader (the industry leader), Coca-Cola could have been purchased with as little as a 25% Margin Of Safety (discount).
At a 25% discount to value, Coca-Cola could have been purchased at any time at or below $44.37. In 1988, the company's stock traded between $35 and $45.25, giving Buffett a discount between 24% and 41%.
Today, Buffett's stock in Coca-Cola is worth more than $10 billion, and he collects more than $270 million a year in dividends. Not bad, considering how easy it was to find the value in this "no-brainer" investment.
Wall Street thought Buffett was nuts. In 1987, earnings were down nearly 2% from their 1986 peak—surely not the sign of a growing company! With a price-to-earnings (PE) ratio of 14 to 19, the company seemed fairly valued at best, if not overvalued.
And once again, Buffett showed the world why Wall Street's earnings mean nothing to the business investor, how to invest like a business owner, and why you are right when your data and reasoning are right—not because the crowd agrees or disagrees.
A quick thanks to Chris at MSU for finding the annual reports and making my job easy!
Please wait while your comment is submitted. (It may take a moment.) Comments on F Wall Street are moderated which means that your comment will appear only after it has been reviewed by Joe. Comments are typically reviewed and approved (or denied) quickly, except between 11:30PM and 5:00AM (CST) – Joe has to sleep some time!
Thank you for participating on F Wall Street. Once your comment has been approved, it will appear here. While waiting, check out some other articles on the blog or click here to return to the article.
| Excel 2007 | | | Excel 2003 |
| (ZIP, 168kb) | (ZIP, 138kb) |
Thu @ 3:33PM | View comment
MinorityStakes said,
A couple comments regarding BBEP's latest communication with shareholders:* 2009 production just about equaled 2008 production even though capex was...
BreitBurn Energy: Playing the Commodities Crash
Sun @ 11:09AM | View comment
Eric T said,
Instead of inventory turnover, I use the cash conversion cycle, or CCC.It is more accurate for companies that manufacture and...
Understanding the True Profit Margin
Sun @ 5:48AM | View comment
Diversification said,
well it all depends on the correlation between the stocks you have choosen many big mutual funds are having the...
The Dangers Of Overdiversification
Sun @ 4:46AM | View comment
sandesh trivedi said,
Very well explained joe. i believe one must also take into account the nature of the product being manufactured while...
Understanding the True Profit Margin
Sat @ 10:19AM | View comment
Ron said,
Hi Joe,Is there a rule of thumb of percentage of net shares sold by insiders where we should start to...
When To Watch Out For Insider Selling
Sat @ 10:18AM | View comment
jan said,
joe, any thoughts on jackson hewitt? what were the risks that played out in your mind when you decided...
BreitBurn Energy: Playing the Commodities Crash
sridhar
Jul 18th, 2007
5 comments
I looked at the FCF for the last 3 years on Edgar online for for BAC, Citi and GS. BAC has negative FCF for 2004 and 2005 and GS and Citi have negative FCF for the last 3 years. Is there a different way to interpret financial stocks?
Cheers
Sridhar
( REPLY | PERMALINK )
Joe Ponzio
Jul 18th, 2007
Joe on twitter
Ponzio Capital
If FCF is negative, you need to dig a bit deeper to find out why (if you want). What would concern me is a lack of consistency. How can you possibly expect to rely on your value calculation if the past was a roller coaster ride?
Financial stocks, or companies like Walmart that plow all of their FCF into new stores, require more work than normal because you have to figure out how much cash they could generate if they relied solely on their operations and didn't try to aggresively grow.
Investing doesn't have to be that difficult. There are plenty of simple companies that may be worthy of your time and money. However, should you choose to dig deeper, you can also uncover some real gems that others may be missing.
( REPLY | PERMALINK )
Olivier
Jul 20th, 2007
4 comments
Thank you very much for your answer regarding the CROIC.
I have an other question regarding the calculation of the multi-year performance of this CROIC :
For 1978-1985, I take the following formula :
(1.0756*1.0579*1.1015*1.095*1.0737*1.0902*1.1138)^(1/7)
-1=8.7%
that is not the same as your result of 9
I am not sure to have the right formula as I have also different value for the other ones.
Could you please help me ?
Olivier
PS : your blog is very interesting with very good explanations and I am quite impatient to buy your book
( REPLY | PERMALINK )
Joe Ponzio
Jul 23rd, 2007
Joe on twitter
Ponzio Capital
When it comes to CROIC, we look for median annual performances - not the growth of CROIC. CROIC is not a growth rate, rather it is a measure of management performance and we want to consistently see it high. For this calculation, you need to get the median of years 1979 through 1985. Let me explain:
When looking for growth rates from, say, 1978 to 1985, we use the end of 1978 as a starting figure and the end of 1985 as an ending figure. That will give us the growth rate for 7 full years in business - from day 1 of 1978 to day 365 of 1985. In essence, we want to see how quickly it grew during those 7 years in business.
When analyzing the CROIC of a company, we want to know how management performed during those five years - 1979, 1980, 1981, 1982, 1983, 1984, and 1985. To do that, we need the median of those five years, using day 1 of 1979 through day 365 of 1985. If we included 1978, it would factor in the entire year's CROIC from 1978 as well.
Because CROIC is not a growth rate but a performance measure, CROIC resets to zero each year when free cash flow is zero and there is no need to use a prior year's CROIC to calculate this year's CROIC.
Try calculating the MEDIAN using years 1979 through 1985 and see if you get the 9.0%. Let me know.
( REPLY | PERMALINK )
Olivier
Jul 23rd, 2007
4 comments
Of course, you're right. Now I find the correct values.
Thanks for your quick answer.
Olivier
( REPLY | PERMALINK )
Tien
Aug 16th, 2007
hope this following question does not make you bored.
i'm having problem in finding Free Cash Flow 1978-1985 a figure of 17.9%
as the FCF from 1978-1985 were:
139
119
238
243
265
346
448
440
thus the growth rate year after year is
-14.4%
100%
2.1%
9.1%
30.6%
29.5%
-1.8%
add them all up and divided by 7 = 22.16%
which is different from your finding of 17.9%
can you please see what went wrong?
cheers
( REPLY | PERMALINK )
Joe Ponzio
Aug 16th, 2007
Joe on twitter
Ponzio Capital
You took the average of the 7 years, but that doesn't necessarily give you a good picture of how your company will perform through good and bad cycles. I prefer to use various timeframes throughout the ten years to compare how the business can sustain or grow in its various natural cycles.
Download the spreadsheet on the Johnson & Johnson valuation to get a better idea of how I use various timeframes.
To answer your other question about POSCO, I know nothing about the steel industry so any analysis would likely end with me saying, "I don't know." There is (I believe) enough information on F Wall Street for you to figure out how to do it on your own if you want to.
Hope that helps.
( REPLY | PERMALINK )
Tien
Aug 17th, 2007
5 comments
ah..i see, now i understand how you calculate that figure, thanks once again
yes, i'm following (and learning) through articles here and hopefully i can come out with an idea about how to value companies by my own one day
( REPLY | PERMALINK )
Buddy Caviness
Nov 23rd, 2007
1 comment
( REPLY | PERMALINK )
Edward
Dec 6th, 2007
5 comments
Makes me think something is wrong with the numbers. Has anyone else done the JNJ analysis on KO ?
( REPLY | PERMALINK )
Night
Dec 6th, 2007
72 comments
Where it has the "Total" value, or one of those cells right under it, it takes the companies projected value and divides it by some number that was similar to 2890000000 (2.89Bil), which is about the number of shares JNJ has out(2.86bil). So just correct that to the number of shares KO has out, and I think you get the correct number for KO.
KO has 2310000000(2.31Bil) shares out.. hmm I don't think that correction is going to account for such a huge difference. Though, maybe they are over-priced on the market right now; they're trading at a P/E of 27 which is pretty high.
I don't have the spreadsheets infront of me, as I am reformatting my computer and the information is on backup media at the moment..
Someone help please! :)
( REPLY | PERMALINK )
Edward
Dec 6th, 2007
5 comments
http://www.fwallstreet.com/blog/17.htm
( REPLY | PERMALINK )
Reba Wesson
Jan 25th, 2008
1 comment
( REPLY | PERMALINK )
Your Name
Mar 12th, 2010