In recent years as some Japanese companies have moved to issue nonvoting shares, considerable attention has been placed on whether the issuance of nonvoting shares should be regulated, and how that regulation would occur. In this column, I would like to consider this question by focusing primarily on the relationship between the allocation of voting rights among shareholders and the structure of corporate control. Specifically, what impact do nonvoting shares have on the corporate control structure?
One-share-one-vote principle
At a general shareholders' meeting, shareholders decide on important ownership matters including the approval of the board of directors and merger agreements. In principle, such decisions are made by a majority of votes exercised by shareholders. Therefore, decisions made at general shareholders' meetings are significantly affected by the allocation of voting rights among shareholders. How does the Companies Act regulate this?
The law stipulates the following: "Shareholders...shall be entitled to one vote for each one share they hold at the shareholders meeting." This is doctrinally interpreted as providing for the one-share-one-vote principle, which is the fundamental rule governing the allocation of voting rights among shareholders. Essentially, the one-share-one-vote principle comprises three key elements. First, anyone who has become a shareholder by acquiring a share in a company is entitled to exercise his or her voting right in the company's general shareholders' meeting (hereinafter referred to as Principle 1). Second, the number of voting rights a shareholder is entitled to exercise is proportional to the number of shares held by the shareholder (Principle 2). Third, there exists a value judgment that as a general principle underlying the preceding two principles, the number of voting rights individual shareholders are entitled to exercise should be proportional to the risk associated with the company's business activities that has been assumed by the respective shareholders (Principle 3).
The Companies Act does not provide for any exception to Principle 2. For instance, it is not permissible to allow a certain shareholder to exercise 10 voting rights per share for each share that he owns. On the other hand, certain exceptions to Principle 1 are permitted. That is, the law allows companies to issue shares with different voting rights. One leading example is the issuance of nonvoting preference shares by Ito En, Ltd. in 2007. Investors holding nonvoting preference shares in Ito En are entitled to extra dividends on top of the regular amount paid per share of other classes of stock issued by the company. However, as long as the dividends are paid as promised, shareholders holding these nonvoting shares cannot vote in general shareholders' meetings (See Ito En's news release "Re: Gratis Issue of Preference Shares and the Terms and Conditions Thereof" http://www.itoen.co.jp/ir/class1/outline/index.html (Japanese)).
Why do companies issue nonvoting shares?
How do companies benefit from issuing nonvoting shares? Ito En, which has already issued nonvoting shares, and many other companies that reportedly have it under consideration have cited "diversifying the means of raising funds" and "securing financing with greater flexibility" as two of the primary reasons for issuing nonvoting shares (See Usen Corp.'s news release "Re: Amendments to the Articles of Incorporation" http://www.usen.com/admin/corp/news/pdf/2008/081104_1.pdf (Japanese - PDF:71KB)).
Indeed, the issuance of nonvoting preference shares - i.e., nonvoting shares with preferred rights to dividends - would provide investors with moderate-risk, moderate-return investment opportunities as a midway alternative between corporate bonds and common stocks. However, the greatest advantage to issuing nonvoting shares for issuers is the ability to raise capital market funds while preserving their existing control structure.
For instance, suppose that Corporation A is a listed company with 100 shares of voting stock outstanding, of which 51 shares are held by Shareholder X. Corporation A is therefore controlled by Shareholder X who owns a majority of voting shares. Now, if Corporation A issues an additional 50 shares of voting stock to third parties, Shareholder X's share of the voting rights will decrease from 51% to 34%. However, if the newly issued shares are nonvoting shares, Shareholder X's share of the voting rights will remain unchanged.
The nonvoting shares to be issued by Corporation A in the latter case differ from the voting shares already listed and outstanding, at least in terms of whether or not they carry voting rights. Therefore, from the viewpoint of protecting investors/shareholders, it is necessary to impose special disclosure requirements on companies issuing nonvoting shares at the time of the initial offering as well as in the course of secondary trading. Furthermore, the issuance of nonvoting shares by Corporation A in the above example would be problematic in that it would reinforce Shareholder X's dominant control over the management of Corporation A. By taking advantage of nonvoting shares, Shareholder X would be able to raise funds without worrying about undermining his control of the company. In this context, the issuance of nonvoting shares can also serve as an effective defense against hostile takeovers.
From the viewpoint of fixed control structure, a more troubling problem could occur because the issuance of nonvoting shares could create a conflict of interest capable of distorting a controlling shareholder's incentives for exercising voting rights and control over the company. If we return to the previous example, Corporation A has a total of 150 shares outstanding - 100 shares with voting rights and 50 shares without. With all other conditions kept equal, the risks associated with business activities of Corporation A are allocated almost equally between Shareholder X, the other voting shareholders not including Shareholder X, and the nonvoting shareholders, at 51, 49, and 50 units of risk (shares), respectively. When the number of nonvoting shares further increases, the share of risks assumed by nonvoting shareholders would increase again. Nevertheless, they are not entitled to voting rights proportionate to the risks they are exposed to. Corporation A would remain under the control of Shareholder X bearing a smaller proportion of risk than the block of nonvoting shareholders.
Under this situation, Shareholder X may exercise his control over Corporation A for purposes other than maximizing shareholder interests or corporate benefits. For instance, assume that Shareholder X increases Corporation A's benefits by a value of 100 through a proper exercise of his power of control. In this case, Shareholder X would benefit in the form of either an increase in dividends or a rise in the share price of Corporation A. But this act would also benefit other shareholders just as much, i.e., in proportion to their respective shareholding ratios.
On the other hand, what if Shareholder X were to enter into a transaction between himself, as an individual, and Corporation A, such as the sale of some of his personal assets to the corporation. If the terms of this transaction come out favorable for him on the individual side of the deal, Shareholder X would not have to share any benefits with the other shareholders of Corporation A. Of course if such a transaction results in a decrease in Corporation A's value, Shareholder X would suffer damage as a shareholder. But if the potential benefits of self-dealing by Shareholder X as a counterparty to Corporation A are great enough, it creates a conflict of interest and increases the possibility for Shareholder X to act on his own behalf at the expense of his fellow shareholders in Corporation A.
However, it is reasonable to grant Shareholder X a disproportionately large number of voting rights relative to his share of business risks in the case where either Shareholder X is the founder of Corporation A and indispensable to the success of its business activities, or in a situation where Corporation A is in trouble and Shareholder X is financing its restructuring. In each one of these scenarios, an extraordinary factor exists that overrides the controlling shareholder's incentive to act on his own behalf at the expense of the company. I believe that such relationships between Shareholder X and Corporation A - i.e., relationships that exist outside the ordinary shareholder-corporation relationship - should be evaluated when considering the advantages and disadvantages of a fixed control structure.
How should the issuance of nonvoting shares be regulated?
There are both advantages and disadvantages to cementing the control structure of a company. One important role of the Companies Act should be to prevent companies from adopting a control structure that poses significant disadvantages while at the same time promoting the adoption of a structure with greater advantages.
In regard to the issuance of nonvoting shares as in the case of Ito En, the Companies Act simply put a ceiling on the number of such shares that companies were allowed to issue. More specifically, the number of shares with restricted voting rights may not exceed one half of the total number of shares outstanding. Such a restriction is meaningful to the extent that it prevents the aforementioned problem of distorted incentives. However, it is difficult to address needs arising from increasingly diverse terms and conditions - those other than voting rights - attached to different share-classes of stock or increasingly diverse risk allocations among shareholders.
Playing a crucial role in the regulation of nonvoting shares under the current Japanese legal framework has been the Tokyo Stock Exchange (TSE). The TSE does not prohibit listed companies, or applicants for listing, from issuing nonvoting shares, but it does impose both procedural and substantive regulations on issues. For example, both listed issuers and prospective listed issues are required to adopt a scheme that would effectively resolve any structure in which a company is placed under the control of a shareholder holding an inordinately small equity stake (Sections II 6 (4), III 5 (5), and III-2 of the TSE guidelines for listing examinations).
The control structure problem addressed by the TSE guidelines is nothing more than the aforementioned problem of distortions in a controlling shareholder's incentive to exercise voting rights or controlling power. Compared to the restriction placed on the number of nonvoting shares, a type of regulation that tends to become unyielding, the provisions under the TSE guidelines are reasonable in that they provide room for consideration of specific circumstances under which a company may issue shares with restricted voting rights.
It may be premature to assess the adequacy of the TSE provisions at this stage because there has yet to be an actual case where these particular provisions have been applied. In regulating the issuance of nonvoting shares or any other shares that do not comply with the one-share-one-vote principle, it would be necessary to determine empirically how corporate management could be affected by distortions in a controlling shareholder's incentives for exercising voting rights or controlling power. Specifically in the case of Japan, it is necessary to analyze the costs and benefits of imposing regulations in addition to the existing restrictions under the Companies Act that limit the number of nonvoting shares that can be issued.
Before determining whether or not such regulations are truly necessary, it is important to first determine if any actual instances have occurred where investors and/or market intermediaries such as securities firms have be able to screen out undesirable nonvoting shares - i.e., those that are potentially detrimental to investors. Clearly more research needs to be conducted in this area before we can develop full understanding of the impact of nonvoting shares on listed companies.
