So I started thinking about how Warren Buffett measures his investment performance – by comparing the change in book value of Berkshire Hathaway to the total return of the S&P 500.
…and I thought, that isn’t fair. It seems too easy to game!
Here’s how I’d do it. Assume it’s January 1, 2013 and I have $10,000. All I need to do is invest in a single high ROE stock and it should deliver a large change in book value almost by definition. I head over to Value Line, check out the free reports for the Dow 30 stocks, and download the one for Coca-Cola (KO).
Let me reproduce the important numbers for KO below:
We start 2013 with BV/S = $7.34. Value Line predicts it’ll end the year at $8.05. When they pay that dividend, it goes into my bank account.
So if I consider myself a holding company, and put my $10,000 all in KO today, I expect at year end to have changed my book value by:
($8.05 + $1.12) / $7.34 – 1 = 25%
Just let the S&P try to beat that!
But… something’s bothering me. That seemed too easy.
Plus, nowhere did the price I paid come into the picture. Since KO was at a P/B = 5 when I bought it, did I just trade $10,000 in book value of cash for $2,000 in book value of KO? That would be an 80% loss!
Snippets of the dreaded financial statement analysis from the CFA Program start to come back to me. Something like this happens when one firm acquires another:
(1) You value their net assets at market value
(2) You add those net assets to your balance sheet
(3) You also add a goodwill asset
And that indefinite goodwill line item is the premium you paid over their net assets – a present reminder of past transgressions!
So let’s go back and redo the math. And assume KO’s net assets are being carried at fair value.
Since KO was at P/B = 5 when I bought it, my balance sheet went from:
$10,000 in cash
$2,000 in KO net assets
If dividends and book value indeed deliver +25%, at year end I will have:
$2,500 in KO net assets and dividend cash
So I changed my book value from $10,000 to $10,500. A meager 5%!
You just can’t get away from being careful about what multiple you pay, even for high ROE franchises.
A reader pointed out that the above accounting isn’t correct (or allowed). What I was going after is a sort of look-through increase in my book value. The way the numbers should be handled on real accounting statements depends upon a couple factors – most importantly, how much control over the company you gain through your purchase.
Fair Value Method:
If the % of KO we buy is small, such that we don’t acquire any significant influence over the company, and we’re planning to hold the position for at least a year or more, we would classify the position as Available for Sale, and use the Fair Value Method of accounting. The net result is that the increase in book value at the end of the year is the cash dividend we collect plus the change in market value of the position. Since we depend on the stock price at year’s end, we can’t really predict what our change in book value will be.
If our position is larger – say 20-50% of KO such that we now have significant (but not total) influence over the company, we’d use this method. Here, we’d record our initial investment in KO as the $10,000 price paid for roughly 270 shares. Value Line predicts we’ll get back 270 x $1.12 = $302.40 in dividends and our line item for our investment will change by 270 x ($2.15 – $1.12) = $278.10. Thus the total change in our book value is $580.50 or 5.8%.
I’ll skip methods for full business combinations (mergers, acquisitions). But since the balance sheets combine, I believe the net result would be similar to the equity method.
It’s my understanding that Buffett actually pays attention to increase in look-through earnings rather than increase in look-through book value.