A nagging feeling came over me while finishing this latest
book with the help of my friend Scott Thompson. While I was happy
with our new book "1988 valuation of Coca-Cola: Estimated
Intrinsic Value," I felt like we did not fully explain the
importance, value and power of that bargain purchase. How could a
man who wrote a book on Buffett and Munger's "Four Filters
Invention" investing process, "Price to Value" and "MOATS," fail
to understand the power of this purchase?

I started emailing friends and doing simple, somewhat crazy math
estimations to see if I could find the truth myself. I am even
embarrassed to say that I sent an email to Mr. Buffett with
flawed math. Friends, with their good intentions, would tell me
to return to "Time Value of Money" calculations, and learn those
well. They would say something like, "Coca-Cola is a great
company with nice rate of returns and steadily increasing
dividends, but it only gives you a range of 9% to 11% returns."

I would think, "How can that be? My investing heroes,
Warren Buffett
and Charlie Munger" always smile and resist cries from
shareholders to sell the the Coca-Cola (
KO
) shares. What I rediscovered is that the magic of
Warren Buffett
's and Charlie Munger's 1988 purchase of Coca-Cola stock is:
yield on cost.

Think of yield on cost" as "yield relative to my cost" or "the
yield received per share cost" or "yield/per share cost." It can
be described as "yield/cost per share."

I sort of backed into finding that bond concept. My initial
thinking went something like this: Think of it as getting an
average of a $570 million dividend every single year from that
principal investment cost of $1.299 billion in Coca-Cola. If you
multiply $570 million x 24 years, we get $13.15 billion. Add this
$13.15 billion to the $1.29 billion principal, and we get very,
very close to the current value of $14.44 billion in BRK's
ownership of Coca-Cola.

Bear with me on this basic "non-compounding" math below and you
will see how Buffett received about $570 million for every year
of that his principal investment of $1.299 billion in Coca-Cola.
Keep in mind, I was trying to stay with simple math. Look at
their cost of $1.299 billion, 0.44 rate of average annual return,
and 0.57, which represents my assumption of $570 million.

0.57 x 23 (23 because of end-of-year adjustment), plus one-half
of year of approximately 0.29, so 1.299 + 13.43 = 14.73, is darn
close.

I found an old 2010 article that said: When Buffett began
purchasing stock in Coca Cola in 1988, many Wall Street analysts
were skeptical because it seemed only a matter of time before
other beverage companies would take away its market share. In
addition, Coca-Cola had reported earnings down 2 percent from the
previous year, and had an unimpressive P/E ratio of between 14 to
19. At the time, shares of KO were worth between $35 and $45. The
stock has split three times since then, and is now priced in the
$60 range. By 1995, Buffett owned 100,000 shares of the company
with a cost basis of roughly $1.2 billion. As of September 2010,
Buffett's unrealized gains on KO were $10.4 billion. This comes
out to a 766 percent increase in value. This is one of Buffett's
greatest investing triumphs.

Using the same simple logic... When $2 grows into $6, no matter
the duration, we say 6/2=3 and 3*100 = a 300% increase in value,
no matter if it takes 1 or 900 years. In this simple math,
duration is irrelevant.

Now, by 2013, KO stock had split 4 times and BRK has 400 million
shares with a cost basis of $1.299 billion, and a market value of
$14.5 billion. Forget for a moment that it took about 24.5 years
to get there. Furthermore, suspend the idea of splits because we
know the cost and the present dollar value that is already
split-adjusted.

When $1.299 billion grows to $14.500 billion, we say
14.500/1.299=11.16 and *100 is a 1,116% increase in value. Again,
in simple math, how much can we allocate to each year? Let us use
a simple average and make it even. Now, a simple rough average of
1116/24.5 years=45.55% approximate gain per year, and this does
not even count the value of the dividends.

(Next, I get a little theoretical.) Let us add in the low-ball
but fair figure of $5 billion for all the dividends (with no
major time value of money adjustments). When $1.299 grows to
$19.500, we say 19.500/1.299=15.01 and *100 is a 1,501% increase
in value. Now, a simple rough average of 1501/24.5 years=61.27%
gain in value (on top of the $1.29 billion) per year, or around
$570 million each year.

Now, I felt like I was getting close to why
Buffett
's 1988 bargain purchase of KO is so powerful and important.
Next, I got a nice email from Richard Griebe. Griebe said he was
starting to see the way I was looking at this investment in KO.
"Rather than looking at the compounding of value over time, you
are looking at the average annual increase in value against the
original $1.299 billion invested. So, if I think of the original
stock purchase as buying a bond instead, that "bond" has paid a
continuously increasing interest rate over time. Following your
computations to where you included dividends to calculate an
average 61.27% gain per year or, in my bond model, Buffett bought
a bond for $1.299 billion that has paid on average coupon of
61.27% annually. This is a feat that would make gangsters
jealous. Thanks for patiently discussing this fascinating case
study with me.

With Griebe's positive words, I felt encouraged and I thanked
him. Next, I kept searching the Internet for this "yield" concept
that I was looking for. I was looking for
Buffett
's effective yield per share compared to my yield per share. I
stumbled upon the concept of yield on cost.

WOW! That is it!
**Yield per "share cost."**

Did I realize that 1.299 billion/400 million shares = Buffett's
$3.25 per share cost per share of KO? Did you?

From the website Investopedia, Yield on cost (YOC) is defined as:
"The annual dividend rate of a security divided by the average
cost basis of the investments. It shows the dividend yield of the
original investment. If the number of shares owned by the
investor does not change, the yield on cost will increase if the
company increases the dividend it pays to shareholders; otherwise
it will remain the same."

To calculate yield on cost for a stock, an investor must divide
the stock's annual dividend by the average cost basis per share
and multiple the resulting number by 100 (to get a percentage).
For example, an investor who purchased 10 shares of stock at $15
and 20 shares at $18 would have an average cost basis of $17 per
share ($15*10 + $18*20)/(10 + 20). If the annual dividend is
$0.90 per share, the yield on cost would be 5.29% ($0.90/$17 *
100).

Using this information, and knowing that Buffett's cost per share
of KO is $3.25, can I calculate his yield on cost for 2012? The
2012 dividends per share were: March 13, 2013 $0.28, Nov. 28,
2012 $0.26, Sept. 12, 2012 $0.26, June 13, 2012 $0.26, March 13,
2012 $0.26, and the sum is $1.30

So, $1.30 / $3.25 = 0.40 and 0.40 * 100 = 40%. Buffett and
Berkshire Hathaway received a 40% yield on cost just for the year
2012 dividends alone!

Alternatively, and again thinking in bond-like thoughts, if we
believe the $570 million average return per year on top of the
$1.299 billion principal. $570 million / 400 million shares is
1.43, and that is like a gain of 43% each year over the initial
investment.

Prediction: Since 45% + 40% = 85%, I predict that the total
yearly return will soon surpass the initial $1.29 billion cost
basis of this Coca-Cola investment.About GuruFocus: GuruFocus.com
tracks the stocks picks and portfolio holdings of the world's
best investors. This value investing site offers stock screeners
and valuation tools. And publishes daily articles tracking the
latest moves of the world's best investors. GuruFocus also
provides promising stock ideas in 3 monthly newsletters sent to
Premium Members
.