A Brief Response

Forever is a long time — indeed, too long. That is the essence of my answer to my two friends and colleagues — professors Zohar Goshen and Joshua Mitts — who each argue against mandatory sunset provisions on super-voting stock (Professor Gordon provides an overview with which I largely concur). Even if one accepts the Goshen/Mitts premise that the other shareholders want the founder to have total control (in order to pursue his “idiosyncratic vision” for the company), the probability is high that, at some point, the majority of the shareholders will want to limit or end that total control. Obvious reasons include: (1) the founder has gone senile (think of Sumner Redstone); (2) the founder has been disgraced or gone to prison (think of JD.com, a Nasdaq-listed company where the CEO was nearly indicted for rape this year), or (3) the founder has caused the firm’s stock price to sink by over 75 percent because of a disastrous, obstinate decision opposed by everyone else (as happened this year at Snap, Inc. where the CEO insisted on switching to a new app that his staff told him was not ready).

Beyond these reasons, the claim that the shareholders want the founder to have total control does not imply that they want that total control to pass to the founder’s heirs or buyers. In fact, many (but not all) firms limit super-voting rights to the first holder only.

Suppose one wants to protect the founder and his vision from interference by those with a short-term perspective. Even under these circumstances, there are less extreme and less overbroad means to this end than dual class stock. Consider the following alternatives to a permanent class of super-voting stock:

  • A sunset provision that reduces the super-voting stock to one vote per share and that expires at some point after the IPO with a right in the founder to convert his or her super-voting stock into common shares at a ratio that compensates the founder for the loss of the control premium;
  • A sunset provision that would be instead triggered only by the common shareholders’ vote to exercise an option given to them at the time of the IPO, under which the founder would be issued common shares at a similarly favorable ratio (with the super-voting stock being redeemed and retired). This provision would never be triggered so long as the common shareholders were content with the founder’s performance, but it would remain as the shareholders’ backstop protection;
  • Longer staggered boards (possibly for four or five years so that insurgent attacks would be discouraged); or
  • Golden parachutes that would also chill the desire of activists to replace the founder — unless a costly forced exit became desirable.

All these alternatives insulate the founder but do not confer perpetual control. Such compromises protect not only shareholders, but other stakeholders as well (who are even more injured by a corporate collapse). No preference among these is here expressed, but they could all be justified depending on the circumstances.

Finally, in this era of Trump, the case for checks and balances should be clear to all. Imagine how bad things would be today if in 2016 the voters could have elected a president for life (in order to enable him to pursue his special “idiosyncratic vision”). When we shift from public law to private law, the answer does not change: Forever is too long.

This post comes to us from John C. Coffee, Jr., the Adolf A. Berle Professor of Law at Columbia University Law School and Director of its Center on Corporate Governance.

1 Comment

  1. Professor Yvan Allaire

    Professor Coffee,
    It seems that your last comment about Trumpian democracy would imply that you do make an equivalence between “one person-one vote” of political democracies and “one share-one vote” of corporate governance.
    The first and, in the mind of opponents, the most compelling argument reproaches the undemocratic nature of different voting rights. One share-one vote becomes the equivalent of the hallowed “one person-one vote” precept of democratic political system (a recent achievement, it should be pointed out).
    For all its superficial plausibility, the argument is hollow. The equivalent of “one person-one vote” to the domain of shareholding would be “one shareholder-one vote” irrespective of the number of shares a shareholder actually owns. Indeed, in political democracies, citizens do not acquire more voting rights because they pay more taxes to the government.
    The analogy of shareholding with citizenship falters in other respects: short-term “tourist” shareholders should not get to vote for the same reason tourists who happen to be in a country on voting day cannot claim voting rights; and then, “new comers” to the shareholding of a company would have to wait for a significant period of time before acquiring “citizenship” and the right to vote as is the case for immigrants, even though employed and paying taxes. Clearly this argument cannot be taken seriously.

Comments are closed.