Courtesy of Martin Chang*
Issuing multiple classes of stock is a strategy that corporations have used to insulate insiders from outside investor pressure. Each class of stock can differ in a variety of ways, but the most common difference is the number of votes that each share is entitled to. By creating different classes, each with different voting rights, corporations can raise large amounts of money in an IPO while keeping control within a core group of pre-IPO investors, founders, and key employees. Facebook, Alphabet, Viacom, and Berkshire Hathaway are just some of the companies that have employed such a strategy. As an example, Facebook employs a dual-class structure, offering a Class A and a Class B type of stock, each of which offers the same economic rights (also called cash flow rights), meaning that they both pay out the same in the event of a merger or bankruptcy. Class A shares, however, typically owned by company insiders, give their owners ten times the votes (per stock) of Class B shares. Because such arrangements insulate management from outside investor pressure, the practice is criticized for contributing to poor corporate governance. For the remainder of this post, I use the less precise term, dual class, to represent the many varieties of multiple class stock structures that exist.
Pros and Cons
Proponents of dual-class stock argue that such arrangements allow companies to be more focused on long-term success. For example, Mark Zuckerberg recounted that in 2012, just after Facebook went public, the company faced intense pressure to advertise on mobile devices. Instead, Zuckerberg focused on building a better user experience, a move made possible by a long-term focus facilitated by dual-class stock.
Institutional investors are the most influential critics of dual-class structures, and they continue to apply pressure against such arrangements. For example, the Council of Institutional Investors (CII) advocates for a “one-share-one-vote” principle. In 2017, they were instrumental in pressuring the S&P 500 and the Russell 2000 to no longer add dual-class companies to their indices. Pushing in the opposite direction, companies have become bolder in reducing the voting rights of stock sold to the public. In 2017, Snap Inc. held the first IPO in history to exclusively sell stock without voting rights.
Researchers and regulators around the world have debated whether government intervention to restrict the use of dual-class stock is appropriate. Proponents of the free market solution, also called “private ordering,” argue that the risks inherent in dual class structures are the type that the market is equipped to handle. The perceived riskiness of dual-class firms is already accounted for in market valuations.
Proponents of regulation suggest that stock prices do not reflect the full effect of the harm that dual-class companies can have on society and the economy. The influence that large corporations have on society suggests that bad governance should eliminated, even if it has no impact on asset values.
For the sake of brevity, I won’t go into the century-old historical background of this debate, but it can be found in my full article, or in Professor Bainbridge’s thorough history of the corporation as a legal entity. In short, although the “one-share-one-vote” norm has been in place for much of the twentieth century, the debate over alternative arrangements has persisted almost as long as the corporation itself.
The effect of a dual-class structure on the investing public depends in part on market efficiencies, which depends on when in the life of the corporation the dual-class structure is implemented. The lower-vote class (of the dual class structure) is often sold in an initial public offering (IPO), and scholars have argued that IPO’s, especially for tech companies integrated in youth culture (e.g., Facebook), are surrounded by a hoopla of irrational exuberance, resulting in unreasonably high valuations.
The lower tier stock can also be issued after an IPO. When the company brings in additional money through the sale, this issue is called a recapitalization. There can also be non-recapitalization issues, such as when they are distributed as a dividend. For example, Google issued its Class C stock, with no voting rights, as a dividend to its existing Class A owners, with one vote per share, as part of its stock split.
Studies suggest that dual-class shares are not, on average, overpriced in an IPO: single and dual-class firms show no difference in average performance for up to three years after the offering. In addition, prices reflect an understanding that risks are associated with shielding from outside pressure. Dual class companies trade at lower valuations than similarly situated single class companies. The effect of such insulating arrangements can have different effects on managers of different temperaments. For example, a CEO such as Warren Buffett may not need much outside pressure to successfully return value to shareholders, given his long track record.
In contrast, various mechanisms may conspire to prop up an IPO price. Company insiders are often contractually bound to keep (i.e. not sell) their stock during a lock-up period after an IPO. Before the lock-up period expires, bearish investors will be hard pressed to find stocks to borrow and short (thus driving down prices). When the lock-up period ended for LinkedIn stock, its price fell 7%, thereby providing anecdotal evidence for inefficient pricing.
These studies all indicate that market inefficiency (in terms of stock prices) is the least convincing factor in favor of dual-class regulation.
Other Considerations (Not Related to Economic Efficiency)
Another consequence of a ban on dual class structures is that founders—fearing a loss of control—will keep their companies private, and perhaps under-funded, to the detriment of society and public investors. Proponents of regulation respond by noting that the market cannot identify, in the long-term, which managers need more public oversight than others. And even if poor performance by managers can be predicted, giving them access to capital, even at a steep discount, is a detriment to society and the economy as a whole.
If there is a problem with dual class structures, there is no shortage of possible solutions. Most solutions focus on the issue of economic efficiency, but society—and by extension, the government—prioritizes other values, such as taking care of the poor, the weak, and the old. I consider this view near the end of the post.
Ban on Dual-Class Structures
As a practical matter, a ban on dual-class structures would likely have to come in the form of state law. Commentators generally find that the Securities Exchange Act of 1934 grants the Securities and Exchange Commission (SEC) the authority to establish rules about corporate disclosures, but not substantive governance issues. Specifically, the D.C. Circuit decided in Business Roundtable v. SEC, that the SEC’s Rule 19c-4, preventing companies with super-voting shares from being listed on national exchanges, exceeded the agency’s authority to make it.
Outside of the United States, countries such as Germany, Poland, and Spain have laws limiting the use of dual-class structures. A country-to-country study of economic performance as a result of governance requirements would be useful, although it would be difficult to isolate the effects of regulation of stock classes.
Professors Bebchuck and Kastiel have argued for a mandatory sunsetting of dual-class structures, in order to gain the benefits of visionary, forward thinking by company founders for a certain time (with ten years being a good candidate), but to avoid the disadvantages mentioned above after the companies have matured. Companies have to be managed differently when they grow from lean startups to big companies with more complex bureaucracies. And founders’ abilities to stay innovative can decline with age. Their proposal would allow the dual structure to renew for another ten years if a majority of the non-voting shareholders vote for it. Such sunsetting plans have previously been implemented by some corporations through their charters, and it is not unheard of that shareholders keep the dual class structure when management is performing well. For example, Canadian company Fairfax put to its shareholders a vote to extend in time its dual-class structure, and it was approved by 68.4% of the votes cast by shareholders unaffiliated with the controlling shareholder.
Bebchuk and Kastiel address the objections put forth by private-ordering proponents by saying that the markets are inefficient at processing the riskiness of dual classes, especially during IPO’s. They state, based on general psychological principles, that investors show bounded rationality and attention during an IPO. With all the other information to consider, they don’t pay enough attention to, or value correctly, issues such as anti-takeover provisions in the charter or the loss in value of non-voting stock in the far future when management can no longer respond to the current business exigencies. The only way to address this market failure is with regulation, as in the form of an SEC rule.
What Bebchuk and Kastiel are saying, in essence, is that predictive signals of management ability are not available for a time horizon of longer than ten years, and even if they are, bounded rationality in the market means that prices do not reflect those signals. As discussed earlier, the data is equivocal at best on whether IPO valuations are mispriced. Also, research indicates that markets are constantly getting more efficient. Prices are more accurately reflecting the available public information; exploitable patterns are getting harder to find. Share prices now move more on the day of earnings announcements than in the past, suggesting that new information is absorbed by the market more quickly. If bounded rationality is not bumping up the price of non-voting IPO issues, the authors’ arguments lose a lot of force. Still, if they are right that manager signaling can only forecast for up to ten years the need for public oversight (or lack thereof), then there should be no loss to the economy of banning longer dual-class structures.
Private ordering solution lets market participants decide on the best solution. The market consists not only of companies and investors, but also national exchanges and advisory groups. Investors themselves have many levels of sophistication, including activist hedge funds, institutional investors, and individual retail investors.
One of the most effective, but drastic, examples of private ordering in action is the corporate takeover. Professor Bainbridge argues that proscribing dual class structures in order to limit its anti-takeover effect “puts one a slippery slope indeed.” Dual-class structures are not so distinguishable, in effect, from other schemes routinely approved by courts, such as poison pill provisions. Bainbridge also points to studies to support his conclusions. One such study looked to see if existing shareholders were harmed by dual-class recapitalizations. It found that stock prices did not drop after recapitalization. Other studies purport to have similar findings.
Bainbridge also found that non-voting shares are normally paid the same amount as voting shares upon merger. This indicates that insiders are not in general looking to cash out at a higher rate than non-voting stockholders, due to the premium for control. Indeed, charter provisions can specify that liquidation values of voting and non-voting stock are to be the same (case in point, Google’s corporate charter).
Proponents of private ordering (at least in regard to dual class structures) also cite international trends towards self-regulation. For example, the Securities and Futures Commission of Hong Kong has announced that it will no longer take an active role in regulating IPOs on the Hong Kong stock exchange, which paved the way for Alibaba to issue its dual class shares there.
Finally, something may be learned by looking at the premium investors pay for greater control. The higher voting stocks that confer such control often go by the name Class A and are often limited by provisions that prevent full alienability. For example, Class A stocks may convert to (lower vote) Class B stocks if they are ever sold. Some comparisons, however, are available. As of 2017, GOOGL, the ticker symbol for Google stock with one vote each, traded at a 1.7% premium over GOOG, the stock without voting rights. This premium is in line with premiums in other industries, suggesting that there is a rational mechanism by which a premium for control (or conversely, a discount for giving up control) is priced.
Stock market efficiency is just one way of measuring the worthiness of a dual-class policy proposal. Overall economic efficiency—resources being allocated with as little waste as possible – has some link to stock market efficiency, but it is not the same. Economic efficiency, however, is harder to measure. It is even harder to measure the distribution of benefits that various policy proposals have on different strata of society. Decisions made by corporate managers affect society in almost every conceivable way. Does their insulation from the investing public, as allowed through dual class stock structures, lead to—in some sense—a democratic deficit? While further study is warranted, the evidence suggests that enough market participants—each with competing interests—are involved to ensure that corporate management does not go unchecked.
Other Market Participants
Even without pressure from investors owning low-voting Class B (or even Class C) shares, corporate management still has underwriting banks, venture capitalists, and creditors to answer to. Venture Capitalists (VCs) often exit companies during an IPO, so there is some concern that their interests—even if they align with those of the investing public, do not translate to long-term protection against risks. Some VCs, however, retain control post-IPO, often through preferred stock (sometimes convertible to common stock) and seats on the board of directors. Founders have been known to be fired pre-IPO by insider VCs (e.g., Cisco’s founders); the founders that last until IPO may be better qualified to run a large corporation.
Even within a private ordering solution, the government—specifically, the judiciary—still has an important role to play in the regulation of dual-class stock structures. Courts enforce the duties of care and loyalty owed by directors and officers to their corporations. Should shareholders with second-class stocks (in terms of voting rights) be given second-class treatment in terms of duties owed to them as well?
* Martin Chang is an attorney and graduate of the Georgia State University College of Law. His practice areas include helping corporations with structuring and regulatory filings. You can reach him through his website at: www.martinchanglaw.com. He thanks Professor Anne Tucker for many helpful discussions that inspired the topic of this post.