The debate over dual class firms has morphed from an objection to their very legitimacy to a demand to subject them to a mandatory sunset provision. My colleague and friend, Professor John Coffee, believes that dual class firms are undesirable and should be restricted, but, to his credit, he exposes the problems with mandating the sunset and suggests ways for improvement. Here, I wish to explain why restricting dual class firms might be costly to the economy.
The objection to dual class firms is familiar. In dual class firms, managers hold incontestable control through high voting shares, with a much smaller holding of equity. This is considered a weak governance structure, and “weak governance leads to weak performance.” Therefore, dual class firms should be restricted or at least limited in duration by a mandatory sunset provision. This reasoning, however, is flawed.
Suppose you are presented with three vehicles—a truck, a car, and a motorcycle—and then asked, “Which vehicle is the safest?” You will answer, “The truck.” Then you are asked, “Which one is the best?” You will reply, “Best for what? Moving cargo? Going on a family trip? Avoiding traffic?” If you are told that the answer should have been “the truck,” because “the safest vehicle is the best vehicle,” you will recognize that this answer is flawed. Vehicles trade off safety for functionality, and thus different vehicles will be best for different tasks. Motorcycles are not as safe as trucks, but are great for traveling cross-country, avoiding traffic, and saving on gas. They are also more prone to accidents and might be used for robbing a bank. But few would suggest prohibiting motorcycles, as we recognize the obvious trade-off between safety and functionality. Yet, this is exactly the type of arguments made against dual-class firms.
The allocation of control between shareholders and management also trades off safety for functionality. The manager wants control to pursue her idiosyncratic vision even against shareholders’ differing opinions[1] or to avoid shareholders’ principal costs[2] such as short-termism. As control determines the freedom to manage, it affects functionality. Shareholders, by contrast, value control because it allows them to minimize managerial agency costs by disciplining disloyal or incompetent managers. As control protects shareholders against agency costs, it affects safety. On the one hand, the more control is allocated to shareholders, the lower is the exposure to managerial agency costs (“high safety”), while management has less freedom to pursue its idiosyncratic vision and execute a long-term business strategy (“low functionality”). Such governance is like a truck. On the other hand, the more control is allocated to management, the more freedom it has to pursue its idiosyncratic vision and execute a long-term business strategy (“high functionality”), while the less is shareholders’ protection against managerial agency costs (“low safety”). Such governance is like a motorcycle. This is, indeed, the typical structure of dual class firms.
By owning the majority of voting rights in a dual class firm, the manager controls business decisions and can block any hostile takeover or activist campaign. Given management’s incontestable control, the only remedy left for shareholders against agency costs is suing for breaches of fiduciary duties.[3] However, incontestable control also provides managers with the maximum ability to pursue an idiosyncratic vision and long-term business strategies. Indeed, the fact that prominent technology firms adopted the dual class structure suggests that technology firms might have higher idiosyncratic vision than other firms.
Why then restrict dual class firms? As mentioned above, the reason is that “weak governance leads to weak performance.” This empirical claim, however, is unfounded. It is the equivalent of suggesting that safety does not trade off against functionality, and therefore a motorcycle always functions worse. Once the control trade-off is acknowledged, any governance structure might be good for one type of business strategy but not for others. Indeed, empirical studies exploring the relationship between firm performance and aspects of corporate governance—such as staggered boards, hedge-fund activism, dual class shares—show conflicting results.[4]
A variation of the same claim is that dual class firms inflict more agency costs on shareholders than any benefit gained through pursuing an idiosyncratic vision and long-term business strategies. This is the same as arguing that motorcycles should be banned because they are more frequently used for robbing banks than for leisure activities. Again, the overall empirical findings are inconclusive as to the relationship between corporate governance and corporate performance. Moreover, a different line of studies suggests that preserving idiosyncratic vision might be more important than coping with agency costs. A recent study has found that “the best-performing 4% of listed companies explain the net gain for the entire US stock market since 1926, as other stocks collectively matched Treasury bills.”[5] Even if only some of the firms in that 4 percent had an idiosyncratic vision, then a policy that would lead to their disappearance, in the name of reducing agency costs, would be damaging to the economy.
The current attempt to mandate a seven-year sunset on dual class firms is similarly misguided. The logic for the mandatory sunset has a twist on the typical reasoning: An idiosyncratic vision is only viable during the first seven years. Thereafter, the visionary is “losing it,” or the second generation “does not have it.” But there is nothing in the economy, or in life, suggesting that idiosyncratic vision is timed to the IPO moment and tied to the founder. It can come at any time (Steve Jobs invented the iPhone long after the IPO), and it can be gained by non-founders (Tim Cook created more value after Steve Jobs had gone). It is easy to point to pathological cases (like that of Sumner Redstone) because idiosyncratic vision exists only in a few cases. But losing these few cases may be harmful to the economy. After all, we are a “venture capital society,” we encourage risk-taking and “crazy” innovations, and we acknowledge that only one out of 10 firms will succeed.
In the past, when most shareholders were retail investors, with a few activists and takeovers, managers could pursue idiosyncratic visions and long-term business strategies without fear. Today, with dominant institutional ownership and extensive activism, the most effective tool to insulate managers and allow them the freedom to manage is a dual class structure. There is no reason to limit its use. With many sophisticated parties, the IPO market does not suffer from negotiation failures. Indeed, the effectiveness of negotiations is reflected in the great variety of terms (including many voluntary sunsets),[6] and although increased use of dual class should be expected, still, it is kept below 20 percent of IPOs.
ENDNOTES
[1] For a complete explanation of idiosyncratic vision and its role in corporate governance, see, Zohar Goshen and Assaf Hamdani, Corporate Control and Idiosyncratic Vision, 125 Yale Law Journal 560 (2016).
[2] Shareholders Principal Costs include Competence Costs and Conflicts Costs. See, Zohar Goshen and Richard Squire, Principal Costs: A New Theory for Corporate Law and Governance, 117 Columbia Law Review 767 (2017).
[3] For the trade-off between the use of control rights and litigation, see, Zohar Goshen and Sharon Hannes, The Death of Corporate Law (April 30, 2017). Available at SSRN: https://ssrn.com/abstract=3171023
[4] See, Goshen and Squire, supra note 2, Ch. IV.A.
[5] Hendrik Bessembinder, Do Stocks Outperform Treasury Bills? (May 28, 2018). Journal of Financial Economics (JFE), Forthcoming. Available at SSRN: https://ssrn.com/abstract=2900447 or http://dx.doi.org/10.2139/ssrn.2900447.
[6] Andrew Winden, Sunrise, Sunset: An Empirical and Theoretical Assessment of Dual-Class Stock Structures (August 1, 2017). Available at SSRN: https://ssrn.com/abstract=3001574 or http://dx.doi.org/10.2139/ssrn.3001574.
This post comes to us from Zohar Goshen, Jerome L. Greene Professor of Transactional Law at Columbia Law School.