Square Root Voting: A New Approach to Regulation of Chaebol, Keiretsu, and Other Conglomerate Organizations in Asia

The economies of several important Asian countries are dominated by large business groups. Many of them are family controlled, such as those in South Korea (known as “chaebol”), Israel and India. Others are not, the most notable example of which is the keiretsu of Japan. Whether family controlled or not, these business groups use highly complex webs of cross shareholdings and pyramidal structures to fend off hostile takeovers. Some American academics see them as an attractive model of corporate governance, but locals view them with concern. For example, the Korean families who control the chaebol through a tangled set of institutional relationships are criticized for exercising dominant voting power vastly out of proportion to the size of their economic stake. Other shareholders can thus exert little control over the chaebol, which have outsized political influence.

The recently elected president of South Korea, Park Geun-hye, ran on a platform that included an Economic Democracy agenda that took aim at the chaebol. The Park administration has proposed measures to prohibit new circular shareholding, tighten restrictions on voting rights of shares held by financial affiliates, mandate cumulative voting, and limit the rights of the families in the election of outside directors. The goal of these measures is to reduce the power of the families that control the chaebol.

The chaebol counter that restrictions on equity investments and voting rights by affiliates would force them to use corporate funds for defending against hostile takeovers instead of deploying them for productive purposes such as R&D. They further argue that the true beneficiaries of cumulative voting would be other large (typically foreign) holders, not true small Korean shareholders. The debate mirrors the arguments in the United States about poison pills—voting rules that managers insert in corporate charters that deter hostile takeovers. In both cases, the question is whether restrictions on the ability of management to minimize the influence of outside investors enhance the value of the firm (by eliminating destructive takeover battles) or reduce it (by entrenching management).

Economists tend to be skeptical of the claim that protecting management from competition can be a good thing, but the question is empirical, and the answer is likely to vary from company to company. For that reason, reformers should not condemn chaebol or poison pills in the abstract, but instead demand that firms use voting rules that give all stakeholders the right incentives to maximize the value of companies.

To achieve this goal, we propose a new approach to corporate governance which incorporates a voting system called Square Root Voting (SRV). SRV simply provides that a shareholder who owns n shares can cast  votes. (For example, a shareholder with 1 share has 1 vote, and a shareholder with 100 shares has 10 votes.)

The economic logic of SRV is derived from the theory of market design pioneered by Professors Alvin Roth and Lloyd Shapley, the 2012 Nobel laureates in economics, and is fairly subtle, but we can provide the intuition. When an outside investor proposes a takeover, shareholders will have different assessments as to the likelihood that the takeover will increase the value of the firm, and (given their different holdings) may be affected differently by the takeover. Under a simple one-share-one-vote system of majority rule, large shareholders can exploit minority shareholders by amassing just enough influence to siphon the firm’s profits into other firms or perks that benefit the large holder. SRV ensures that as a shareholder accumulates more shares, his marginal vote becomes more expensive, making it always cheaper for the large holder to pay smaller holders to support his positions and making value-destroying self-dealing impossible.

Furthermore, the quadratic function ensures that shareholders pay a price per vote proportional to the number of votes they buy. As a result, under SRV each investor buys a number of votes proportional to the intensity of his preferences. This results in decisions that maximize the well-being of the group, as Weyl has proven in an academic paper that discusses a closely related voting system called “Quadratic Vote Buying.” When a vote is held whose outcome is more important to the minority than to the majority, the minority can outvote the majority by spending more money on votes, and in this way protect their interests. When the outcome is less important to the minority, the minority will not buy a lot of votes, and then the majority prevails.

In particular, hostile M&As that destroy value will not be approved under SRV, obviating the need for the chaebol to use circular shareholding, mobilize financial affiliates, or use funds for long-term investments to buy its own shares for the purpose of defending against inefficient takeovers.

In addition, SRV allows more effective monitoring by shareholders of the controlling family’s tunneling. For example, major related-party transactions could be put to a vote upon the petition by shareholders with a certain level of combined voting rights. Under SRV, transactions that are expected to create economic value will be approved. Transactions that line the pockets of controlling families at the expense of small shareholders will be rejected.

As a result, SRV will enable the Korean government to do away with both the current and proposed across-the-board, rigid regulations on corporate governance, equity investments and related party transactions, allowing the business tycoons to put their entrepreneurship to work and focus on the long-term management of the conglomerate.

A number of Korean assemblymen have expressed interest in the concept of using SRV to rein in the power of the chaebol without meddlesome government regulations. In Korea, legal changes would be needed to allow firms to use SRV. But in the United States, firms are free to experiment with different voting systems, and managers and directors should be encouraged to incorporate SRV into corporate voting rules. Indeed, the typical poison pill already reflects, in exceeding crude form, the logic of the SRV: it effectively creates a lower price of voting for insiders who own fewer shares than for the outside acquirer who has accumulated a large stake. But this type of poison pill gets the logic all wrong. The cost structure faced by insiders and outsiders should be the same; and the arbitrary thresholds used by poison pills to determine when the special voting rules are invoked, and the arbitrary discounts are hard to defend. A poison pill that simply provided that the SRV voting structure will be used for takeover battles would avoid these problems while ensuring that takeovers take place only if they maximize value for shareholders.