The Distribution of Voting Rights to Shareholders

Vyacheslav Fos is Associate Professor of Finance and Clifford G. Holderness is Professor of Finance at Boston College Carroll School of Management. This post is based on their recent paper.

Our new paper, The Distribution of Voting Rights to Shareholders, is the first comprehensive study of the distribution of voting rights to shareholders. Using over 100,000 distributions of voting rights to shareholders, we find a wide array of evidence that firms and stock exchanges change when they notify investors of the voting record date based on the proposals involved and that sophisticated investors are often notified before retail investors. Trading volume is higher than normal both before and immediately after the record date. Stock prices decline significantly when they go from cum vote to ex vote. These changes in notification, trading volume, and stock prices are correlated both with how controversial votes are and how they ultimately turn out.

The right to vote is one of only three distributions made to shareholders. The other two distributions, cash dividends and rights offers, have been studied for years, with well in excess of 100 papers studying ex day changes with cash dividends alone. Moreover, the most common of the three distributions for most firms is the right to vote because it must occur prior to each shareholder meeting. Finally, voting is central to how shareholders control agency costs and influence key corporate decisions. Our findings show that the distribution of votes is far from straightforward mechanical event.

For votes to be distributed, a firm must first set a record date. Only those who are shareholders of record on that date may vote at the forthcoming meeting. Investors must then be notified of this date. We find that in 91% of the cases firms file a proxy to notify investors of the record date after that date has occurred. This ex post notification stands in sharp contrast to the other major distribution to shareholders, cash dividends, where record dates are inevitably announced well in advance. We further find that whether firms announce the record date in advance is associated with the type of vote before shareholders and the eventual outcome of the vote. Notification of the record date, thus, seems to be one way that managers can influence the voting of their shareholders.

What we found to be even more unexpected is that stock exchange officials also influence the voting process by revealing the voting record date to select investors, often before that date occurs and before the public learns of the date through the filing of a proxy. The New York Stock Exchange (“NYSE”) requires that firms report forthcoming voting record dates as part of its “self-regulation” initiative. The Exchange then sells this information to select investors. These private sales of non-public information, which include other potentially valuable information, are studied here for the first time. They seem to conflict with a core principle of federal securities laws, namely that all investors have equal access to material information.

Even when a proxy has not been filed and there is no exchange notification (because a firm is not listed on the NYSE), we document that some investors somehow learn of a forthcoming voting record date and trade accordingly. Overall, notification of the voting record date in the United States stands in sharp contrast to Europe where by law all investors must be notified at the same time and well in advance of the date itself.

We also document what happens to stock prices and trading volume when votes are distributed, that is when stocks go from cum vote to ex vote. Trading volume is higher than normal before stocks go ex vote. It then declines around the record date, apparently reflecting uncertainties on when trades clear and sellers thus lose the right to vote (an uncertainty not found with cash dividends because exchanges set an explicit ex day). Once stocks have clearly gone ex vote, there can be an immediate surge in trading even though the outcome of the vote has yet to be determined. This surge raises the possibility that some voting shareholders are motivated more by securing private benefits than by increasing firm value, a scenario which has received little attention to date in the academic literature.

To measure what happens to stock prices when they go ex vote, we use the same methodology pioneered by Dolley (1934), to study the distribution of rights to shareholders, and used subsequently by Elton and Gruber (1970) and many others, to study the distribution of cash dividends to shareholders. Manne (1962), in a seminal paper which was one of the first to propose that shareholder voting matters, called for a comprehensive study of this nature to quantify what happens to stock prices when they go ex vote. Surprisingly, ours is the first such study. We find that stock prices typically decline when votes are distributed; that is when stocks go ex vote. The magnitude of this decline varies with how controversial the vote is expected to be and how investors are notified of the record date. For example, the ex day decline averages 66 basis points when proposals by dissident shareholders are involved. We identify instances when stock prices decline by more than 5% when they go ex vote. We interpret these declines as reflecting activist investors buying marginal votes. They are also the losses investors incur by selling stocks ex vote instead of cum vote. Given that retail investors presumably rely on proxy statements to learn of record dates, these losses are often incurred unknowingly because proxies are typically filed after the record date.

Finally, our investigations suggest that different shareholder votes have different underlying dynamics. With some votes that are controversial and close, there are few changes around the ex vote day. This suggests that these votes are decided primarily by buy-and-hold shareholders. But with other votes that are controversial and close, there are marked changes in trading volume and stock prices around the ex vote day. This suggests that these votes are influenced by activist investors who are willing to pay a higher cum vote stock price to influence a forthcoming vote. In general, our investigations highlight the importance of disaggregating shareholder votes.

In summary, while the many empirical regularities we find show that the historic neglect of what happens when stocks go from cum vote ex vote is not warranted, at the same time they raise numerous questions for both policymakers and researchers. Addressing these questions will be important because the ultimate control of any corporation rests with its shareholders and their power comes primarily through voting. Many shareholder votes are perfunctory, but some are not. It is with the distribution of these votes where the ex day changes are the most informative.

The complete paper is available for download here.

Both comments and trackbacks are currently closed.