For many value investors, the apparent cheapness of Berkshire stock is both a tremendous gift but also a somewhat puzzling anomaly. Many highly respected analysts and fund managers have spoken with conviction on this subject, notably Witney Tilson (www.tilsonfunds.com/BRK.pdf) and Donald Yacktman. And, of course, Buffett himself (ipse dixit) has made it clear that he believes the stock is selling at a significant discount to intrinsic value and has authorized share repurchases, a declaration to the investing community that he believes his own stock is one of the best bargains available on a risk-adjusted basis.
The list of potential explanations for such a deviation from intrinsic value are numerous: Buffett's age, the question of leadership succession, the sheer size of the business, etc. However, one under appreciated and possible explanation may be found in Buffett's explicit and irrevocable pledge of a large portion of his Berkshire stock to charitable foundations. This article examines the extent to which such contributions could suppress the stock's price and how an investor might interpret this phenomenon.
In his seminal work “How to Be a Stock Market Genius”, Joel Greenblatt extolls the virtues of the astute “special situations” investor who can identify when certain market participants may, in fact, be compelled to liquidate certain positions irrespective of their underlying value. These “forced sellers” cause an anomaly in the market's pricing system, at least for a short period of time, because their actions are not necessarily governed by the discernment of value versus price-received in a transaction. Could this same situation exist with respect to the charitable foundations who benefit from Berkshire stock donations by Buffett?
Indeed, this is the case to some extent. In 2006, Buffett made five major pledges of stock to private foundations. Each of the five are irrevocable (though some conditions exist for the largest pledge) and follow a basic formula, whereby 5% of the total amount of pledged stock remaining in a given year is given away in that year. The following is a list of the pledges: (note: pledges are in Class B stock, such numbers adjusted and shown below in 2012 Class B equivalent to account for stock split that occurred during the BNSF acquisition).
-Bill and Melinda Gates Foundation (BMG): 500 million
-Susan Thompson Buffett Foundation (STB): 50 million
-Howard G. Buffett Foundation (HGB): 17.5 million
-The Sherwood Foundation (formerly Susan A. Buffett Foundation: SAB): 17.5 million
-NoVo Foundation (Peter Buffett, NoVo): 17.5 million
For a total of 602,500,000 (2012) Class B equivalent shares. See: http://www.berkshirehathaway.com/donate/webdonat.html
Below is a table showing the total pledged shares and the donation schedule of the aggregate amounts to the five entities.
Pledged Stock Remaining, Year End
The most restrictive conditions are on the BMG, in that each year's share donations “must be fully additive”, more fully explained in that “annual giving [of BMG] must be at least equal to the value of [Buffett's] previous year’s gift plus 5% of BMG’s net assets. If this amount is exceeded in any year, however, the excess can be carried forward and be offset against a shortfall in subsequent years. Similarly a shortfall in a given year can be made up in the following year”. In effect, the value of shares Buffett gives to BMG each year must be added to BMG's total giving. It appears that BMG has, to some extent, sold Berkshire shares to meet this requirement. In several years, though, it has sold less than the amount of donated shares.
With respect to the other foundations, it is presumed that the vast majority of their funds are solely the result of the pledge of stock from Buffett. Therefore, the foundations have a strong incentive to sell Berkshire stock to raise cash for their charitable mission. Reviewing the 990's for each of those foundations (source: Guidestar), where available, it appears that they have, to a large extent, sold shares roughly in function to the donation schedule referenced above.
Therefore, we conclude that the charitable foundations are selling Berkshire stock at a rate that is
approximate to (but somewhat lower than) the schedule of total share donations in a given year. If effect, they might not be “forced sellers” in the strictest sense of the word, but they are acting in a way that may be similar to forced sellers. We note that, as charitable foundations, their mission is not investment gain but philanthropy and so maximization of good perhaps trumps maximization of monetary return.
But, this is selling of a material amount? According to Morningstar.com, average daily volume of Class B stock is 4.1 million and Class A volume is 500 (or roughly 750,000 Class B equivalent) for a total Class B equivalent volume of 4.85 million. We compare this to a total annual “volume” of 22 million donated shares in 2011 and we see that these donated-and-sold shares could account for as much as about four and a half trading days worth of volume (assuming all 22 million are, in fact, sold). But, this is hardly a fair comparison! 5 million shares traded daily includes actors such as high-frequency traders that are exploiting minute stock price movements and are holding shares for minutes if not seconds. According to Thompson Reuters, such high-frequency traders account for 60% of the volume of trading.
Other sources push that number to 70% and beyond:
It is impossible to know what portion of the Berkshire share volume is what we might refer to as “real” trading, but it is safe to say it is significantly less than 5 million per day. For the sake of argument, and recognizing the uncertainty of the figure, let us assume that only 50% of the trades are net of high-frequency traders. That would give us a more fair estimate of about 9 trading days worth of “forced-seller-equivalent” supply. Given roughly 250 trading days in a given year, that represents 3.6% of the trading volume (again, after discounting high-frequency traders).
Another way to look at the situation would be to compare the number of “forced seller” shares being sold annually (22 million) to the number of Class B (2012) equivalent shares in existence. As of the 2011 10-K filing, Berkshire stock included 1,072,262,656 shares of Class B stock and 936,053 share of Class A stock. Using a conversion of 1/1,500 for Class A stock, we arrive at a total Class B equivalent share count of about 2.476 billion. Therefore, stock involved in “forced selling” in 2011 could amount to 0.89% of all stock in existences as the end of 2011.
Given these two metrics (with all their flaws and approximations), it appears that the volume of stock
subject to donation in recent years could be having a material impact of the current stock price of Berkshire Hathaway. This effect should be moderated somewhat by value investors spotting a bargain and purchasing shares, but only to the extent that their actions neutralize the forced-seller's sales. If there are not enough value-conscious investors tracking this opportunity, and if they have insufficient resources to “soak up” the excess selling that should not otherwise occur, it seems only natural that the stock price would be suppressed below value. Perhaps this is another reason Buffett, as Chairman of the Board, might be interested in share repurchases: to neutralize the adverse effect on share prices of his charitable contributions of stock. In effect, he gets to both purchase undervalued shares of a great company while also supporting the charitable foundations to whom he has pledged his fortune.
A final encouraging note: unless Buffett were to pass away in the near future, the volume of shares
subject to this “forced seller” effect should diminish gradually according the pledge agreements. So, the enterprising value investor can take heart that the shares are available at a great price and one of the elements contributing to that below-value price should resolve itself in the coming years.
Disclosure: the author holds a long position in Berkshire.