Readers seemed to enjoy my post about anomalies in the pricing of Citigroup securities, so I thought it would be fun to look at another anomaly, this time in the shares of Berkshire Hathaway, Warren Buffett’s famous company.
This pricing anomaly is small compared to the Citigroup one. But it does raises some interesting questions about how well the markets are functioning (and, who knows, maybe someone can profit from it).
The root of the anomaly is that Berkshire Hathaway has two classes of shares: A and B.
The B class was created in 1996 because the price of the regular A shares had risen beyond the reach of many ordinary investors (more than $30,000). Most companies would have solved this problem much earlier by splitting the stock (i.e., by converting each share of stock into multiple shares, each of which would trade at a correspondingly lower price). But the Oracle of Omaha famously eschews stock splits. So Berkshire came up with another plan. It issued new shares, the B class, that would receive 1/30 the economic payoff of the A shares. (For further details, including a difference in voting rights and conversion rights, see the memo from Warren Buffett that’s quoted at the end of this post.)
Now you might think that the price of an A share should be about 30 times the value of a B share — the law of one price in action. And you would be right; the prices have often traded around a 30:1 ratio since the class B shares were introduced. But not always. In recent months, for example, the ratio has increased noticeably, with Class A shares trading at a notable premium to the Class B:
On Friday, the ratio finished the day at 30.8. The A shares closed at $91,600, while the B shares closed at $2,972. At that price, 30 B shares would cost you $89,160, almost $2,500 less than a single A share. If you think that the prices will converge, you might be able to profit by shorting an A share and buying 30 B shares (that’s not a recommendation, and no I haven’t tried).
The question, of course, is why the A shares have decoupled somewhat from the Bs. One possibility is that there are stresses in the financial markets. As the chart indicates, the ratio increased during past periods of financial stress such as the summer of 2007 (when the financial crisis started), the fall of 2008 (after Lehman fell), and the winter of 2009. That makes sense, since those are periods in which arbitrageurs have may had better things to do with their money (e.g., hoard it against liquidity needs) than playing the A-B price spread.
Another possibility, at least in principle, is that investors have decided that the extra voting rights of the A shares are worth paying a premium. Perhaps they know something we don’t?
And perhaps it’s a real anomaly, waiting to be closed by budding arbitrageurs. Hypotheses welcome in the comments.
Here is Warren Buffett’s analysis of the A and B shares (from a memo to shareholders):
Comparison of Berkshire Hathaway Inc. Class A and Class B Common Stock
Berkshire Hathaway Inc. has two classes of common stock designated Class A and Class B. A share of Class B common stock has the rights of 1/30th of a share of Class A common stock except that a Class B share has 1/200th of the voting rights of a Class A share (rather than 1/30th of the vote). Each share of a Class A common stock is convertible at any time, at the holder’s option, into 30 shares of Class B common stock. This conversion privilege does not extend in the opposite direction. That is, holders of Class B shares are not able to convert them into Class A shares. Both Class A & B shareholders are entitled to attend the Berkshire Hathaway Annual Meeting which is held the first Saturday in May.
The Relative Prices of Berkshire Class A and Class B Stock
The Class B can never sell for anything more than a tiny fraction above 1/30th of the price of A. When it rises above 1/30th, arbitrage takes place in which someone ¾ perhaps the NYSE specialist ¾ buys the A and converts it into B. This pushes the prices back into a 1:30 ratio.
On the other hand, the B can sell for less than 1/30th the price of the A since conversion doesn’t go in the reverse direction. All of this was spelled out in the prospectus that accompanied the issuance of the Class B.
When there is more demand for the B (relative to supply) than for the A, the B will sell at roughly 1/30th of the price of A. When there’s a lesser demand, it will fall to a discount.
In my opinion, most of the time, the demand for the B will be such that it will trade at about 1/30th of the price of the A. However, from time to time, a different supply-demand situation will prevail and the B will sell at some discount. In my opinion, again, when the B is at a discount of more than say, 2%, it offers a better buy than the A. When the two are at parity, however, anyone wishing to buy 30 or more B should consider buying A instead.
This is interesting. Our portfolio has both A and B shares. I have always faceteously said the A shares represent ‘smart money’ and B shares ‘dumb money’. Or maybe the A shareholders are more the ‘strong hands’ players and B shareholders ‘weak hand’ players,
I noticed thie pricing vagary recently, as did you. I am non smart enough, or quick enough to act on this. That is what the specialists are charged to do, provide for an orderly market in the security(s). I’m sure the specialist could explain to you how and why the spread varies.
As far as the voting rights [extra voting rights of the A shares are worth paying a premium. Perhaps they know something we don’t] hypothesis, I don’t think that most players purchasing shares on the open market, whether A or B, have enough clout, assuming they know more than the current board, would have any value to effect a board decision.
Good subject matter for a B school study.