We proceed from the assumption that, on the whole, the separation of ownership and control is a good thing. Market actors come by capital in any number of ways (some noble, some neutral- reading Fooled by Randomness makes it apparent that this may well be the lion's share- and some less than noble). It follows then that not all holders of capital will be competent (or even tolerable) managers of enterprises. Outstanding managers may not, consequently, be holders of capital. Making capital a necessary prerequisite to the exercise of managerial skill would seem, therefore, to work a social ill.
The separation of ownership and control, combined with a robust market for corporate control (to promote efficient competition for managerial talent) permits specialization by managers and permits holders of capital to, well, do whatever they like.
Going Private readers, being keen students of the theory of the firm, will quickly recognize the ugly hints of the "principal-agent problem" herein and, when they encounter such governance issues, wonder what mechanisms keep the interests of managers and owners aligned and how well those mechanisms have been priced into the asset being managed.
Going Private readers also scoff at use of the term "moral hazard" in these cases, (the offense of this sort of misuse being committed primarily by certain financial journalists and politicians; the former because "moral hazard" is a much better catch phrase than "principal-agent problem," and the later because the former think "moral hazard" is a much better catch phrase than "principal-agent problem"). Instead, Going Private readers recognize that, while both moral hazards and the principal-agent problem are related to information asymmetry, the use of "moral hazard" as a blanket term here is flawed.
Much hand wringing and many tears have been spilled over defining the appropriate balance between ownership and control, particularly in markets efficient enough to facilitate short-term speculation, which disrupts (not necessarily in a bad way) continuity of ownership and, to the extent long-term ownership yields firm specific experience, reduces ownership knowledge in the aggregate. (The shareholder who remembers 40 years of management teams likely knows more about the underlying firm than the shareholder who purchased the stock last week). Of course, you don't want the old widow of the long dead partner in the firm holding up shareholder meetings to demand the plant be painted pink either. Getting the balance right, goes the theory, is critical to avoid harm befalling the owners. I find this blanket treatment distasteful, and I think with good reason.
Different firms require different balances of ownership and control.
Any number of mechanisms for fine tuning this balance exist (poison pills, super majority voting, staggered boards) and have evolved (with predictable consequences to their complexity) over a long history of battles for corporate control.
Lately, however, dual class share structures have come into the spotlight, primarily, I think, because of the host of newspaper and "old media" firms which- readership dwindling, embattled by the realities of transforming markets and the historically unprecedented (and mostly well deserved) disdain and mistrust of The Fourth Estate by the public at large- look cheap and have therefore come under acquisitive attack.
Dual class share structures take any number of forms but the most common is to issue a "Class A" set of shares and a "Class B" set of shares. The Class A set of shares may, for instance, get 10 votes per share. Class B might, instead, get 1. In this way, unless Class A shares are in short supply, it is highly difficult to overcome Class A voting power. Google uses this structure (to avoid being "evil" I guess, but let's just not mention that whole China thing). The question of who gets Class A and who gets Class B shares at Google (or anywhere else) is left as an exercise for the reader.
Newspapers, to some extent by a trick of fate, also happen to be among the most common firms to employ dual class share structures. Part of me suspects that this is in no small part a consequence of the fact that, because of how newspapers were built after the turn of the century, print media companies tend towards dynastic owners.
Newspapers were, almost by definition, contrarian endeavors tasked with scrutinizing the powerful while themselves dominated by powerful and highly public personalities. Dual class share structures typically concentrated the power in the ruling family and were explained by the need for newspapers to be fiercely independent of the whims of the market (or any other whims for that matter).
I have some sympathy to these structures with media companies, or I used to. Buffett seems to agree with me. He believes in dual class structures for newspapers and has been somewhat vocal about it to boot. Berkshire owns a large chunk of the Washington Post, a dual class structure with the Graham family owning all of the voting shares. Berkshire has dual classes too, but Buffet's power isn't related to any supermajority of voting shares. He holds equal voting weights in all the Berkshire classes and given his plans for his fortune, its hard to say that firm is going to be a dynastic one.
Dual Class structures aren't always a legacy holdover. Sometimes they are used pro-actively to restructure power to the ruling class. About a year ago The Wm. Wrigley Jr. Company issued a special dividend. One share of Class B stock, with 10 votes, for every 4 shares of common with 1 vote. There was a hitch. The voting rights of a Class B share are non-transferable. The shares effectively convert to common if sold. Of course, this means that every time a Class B share is sold, the existing holders of Class B shares enjoy voting accretion. The elegance of this structure is in the self-selection bias it introduces to the shareholders. If they want the money (they are shorter term focused) then they dump the voting rights. (Who wants short term shareholders voting, after all?) If they want the power, then they have to forgo the money.
After the dividend, William Wrigley Jr. had about 28% of the vote. If only a third of the shareholders sold their Class B's he'd hold 40% and the remaining Class B shareholders would, by definition, be believers in management. (Nicely done. I wish I knew who did their legal and corporate finance work). Of course, as a member of "the family" you almost pray for a sharp downturn in the stock price to flush out the fickle, which presents interesting principal-agent problems itself. (I highly recommend "The Hudsucker Proxy," as an amusing diversion in this direction). This was an interesting approach which may well have been driven by mid-1990s rules imposed by the NYSE on dual class structures. Those rules had grandfather clauses for firms that already had such structures in place. (One more reason Google listed on NASDAQ, perhaps).
The New York Times, presently in the spotlight on this issue, has what must be the best self-serving proxy statement on this issue ever written:
The primary objective of the 1997 Trust is to maintain the editorial independence and integrity of the New York Times and to continue it as an independent newspaper, entirely fearless, free of ulterior influence, and unselfishly devoted to the public welfare.
One is led by this to believe that the firm gives its profits to "Jerry's Kids" or something. To shed some more light, the 1997 trust is overseen by eight trustees. All of them are family members. Making a modification to the trust requires the vote of six of the eight trustees. If you were looking for a model to insulate an organization from outside pressure, this is a pretty good place to start.
Dow Jones, owner of, among other things, The Wall Street Journal, is facing similar scrutiny with the absolutely massive premium attached in the form of a $60.00 per share bid (the stock traded at $34.50 just last week) by Rupert Murdoch. If one needs a demonstration of the power of dual share structures, this bid with its 45% premium has a "40% chance of success," in the words of a UBS analyst (for whatever that's worth). Even the risk arbitrage people aren't so sure. As of this writing, the stock hovers at $53.64 per share, a sharp discount from the offer price. It is also illuminating to recognize that the bid comes from another potent media personality.
The Observer demonstrates the pickle comically with a "scratch to win game" that highlights the absurdity of close control in (most) dynastic control structures that have persisted more than a single generational passing of the torch. Look at who you have to win over to prevail. (But then, there are some pretty batty public shareholders out there too, I suppose).
So do dual class structures have a place? Do they even present severe problems? Well, it's clear that they have effects on corporate performance. Masulis, Wang and Xie's "Agency Problems at Dual Class Companies" is but one of a slew of papers examining the issues surrounding dual class share structures. From their abstract:
These findings support the hypothesis that managers with greater control rights in excess of cash-flow rights are prone to waste corporate resources to pursue private benefits at the expense of shareholders. As such, they contribute to our understanding of why firm value is decreasing in the insider control-cash flow rights divergence.
But so what?
I have commented on Cablevision's "Dolan Factor" here before. Specifically, that a lack of confidence in management is priced into a stock's price by rational actors. Also priced in is the potential to change management. There is ample evidence that investors on the margin make buy and sell decisions based in no small part on corporate governance issues. In addition, these issues have gotten a lot more exposure in recent years, particularly with the rise of activism. Pricing a "Dolan Discount" into a stock is par for the course, and that discount doubtlessly gets a multiple if a dual class structure entrenches the Dolans in the firm. This is as it should be. All is well with the market.
Let's remember for a moment that shareholders who buy common in a dual class structure with weighted voting rights to founders or their dynasty got what they paid for. Giving what amounts to free voting rights (and the consequent repricing of the shares to include the value of said voting rights) to shareholders who never had them in the first place is a free lunch. That won't stop activists from using dual class structures as a political weapon, but that doesn't particularly bother me either.
In reality, dual class share structures, at least in newspapers, probably are necessary anymore. The days where the public capital markets were the only place to go to get the kind of money you needed to build or maintain a national newspaper's infrastructure are long past (not to mention that the infrastructure is much cheaper today). Any number of sources, including going private transactions, are available to newspapers today, and doubtless the competent managers in the industry could command significant control for themselves in a private transaction. (Look at what News Corp is willing to pay for
Dow JonesThe Journal).