Although no one can claim to fully understand Berkshire Hathaway (NYSE:BRK-A) (NYSE:BRK-B) Chairman Warren Buffett's genius or how exactly he values a company, it doesn't hurt to try to understand his past purchases. In this case, See's Candy, an extremely simple but successful early acquisition, provides a peek into how the master made a killing in chocolate.

Like other Berkshire investments -- including GEICO, Moody's (NYSE:MCO), American Express (NYSE:AXP), and Inside Value recommendation Coca-Cola (NYSE:KO) -- See's Candy is a cash cow that earns stratospheric returns on capital. From its purchase in 1972 until 1998 (I am relying on a speech made by Buffett in 1998 for my calculations), See's Candy increased pre-tax earnings from $4 million to $60 million, and as Buffett said, "it still doesn't take any [additional] capital."

Is the price right?
Buffett has often remarked that when they bid $25 million for See's, they were not willing to pay one cent more. If the sellers didn't play ball, they would've walked away. In hindsight, that would've been a mistake, but it shows the enormous amount of discipline Buffett and partner Charlie Munger exhibited.

At the time, See's was earning $4 million pre-tax and about $2.5 million after tax, meaning Berkshire (via a future subsidiary, Blue Chip Stamps) paid roughly 6.25 times pre-tax earnings and 10 times after-tax earnings.

Go where the puck is
In order to put a price on See's, Buffett would have had to make some assumptions about See's prospects. One flash of insight Buffett had was that See's customers were extremely loyal and would tolerate annual price increases.

According to Buffett "I bought [See's] in 1972, and every year I have raised prices on December 26th." Thus, even if See's volume growth was slow, the pricing increases would provide the juice to make the investment work over a long period of time.

Thus, if Buffett had been able to gaze into a crystal ball, he would've seen that See's could grow its pre-tax earnings at a rate of almost 11% per year from 1972 to 1998. In 1972, See's made about 25 cents of pre-tax earnings per pound of chocolate sold. By 1998, that profit per pound had increased to $2, or an annual growth rate of roughly 8.3%.

In addition, See's volumes have grown from about 16 million pounds sold per year to 30 million. That's an annual growth rate of about 2.4%. Add them together and you get 11% average annual growth.

The price is definitely right
So what's a company that earns $4 million pre-tax and will grow earnings at roughly 11% for the next couple of decades worth?

To answer that question, we can simply turn to Excel and, with the benefit of perfect hindsight, model in the numbers. If we assume net income equals free cash flow and that the company trades at 15 times earnings in the last year of our model, calculations show that paying $25 million for See's in 1972 would result in a 22% internal rate of return.

Interestingly, that's roughly (give or take a percent) the same rate of return Berkshire's stock and book value per share has grown at since 1972, so clearly See's has performed almost exactly to expectations.

Closing thoughts
In retrospect, See's was a dream come true. It had an extremely wide moat thanks to its loyal customers, it required almost no capital reinvestment, and its pricing power allows it to increase earnings year after year. Fools take note: These are the types of investments that can be decades-long winners.

Related Foolishness:

This article represents the opinion of the writer, who may disagree with the “official” recommendation position of a Motley Fool premium advisory service. We’re motley! Questioning an investing thesis -- even one of our own -- helps us all think critically about investing and make decisions that help us become smarter, happier, and richer.