Print
E-MailTingting Liu is Assistant Professor at Creighton University. This post is based on a recent paper by Professor Liu; Kai Li, W.M. Young Chair in Finance at University of British Columbia Sauder School of Business; and Juan (Julie) Wu, Assistant Professor of Finance at University of Nebraska at Lincoln.
Modern corporations are characterized by the separation of ownership and control, thus shareholder engagement in important corporate decisions is fundamental to the governance process. Despite its importance, evidence on the role of shareholder engagement in one of the most important corporate decisions—mergers and acquisitions (M&As) is limited and mixed. Our paper provides one of the first large sample studies documenting a positive causal effect of shareholder approval on corporate M&As.
In general, it is difficult to find a setting in which a firm’s governance structure changes exogenously. The challenge faced by many empirical studies is the endogeneity of a firm’s governance structure. For example, acquirers whose deals require shareholder approval may be fundamentally different from those whose deals do not require shareholder approval. A simple comparison of these two groups of acquirers only suggests possible association between shareholder approval and deal quality, but does not establish causality.
Our identification strategy—the regression discontinuity (RD) design—relies on listing rules of the NYSE, AMEX, and NASDAQ that shareholder approval is required when an acquirer intends to issue more than 20% new shares to finance a deal. We examine the value effect of shareholder approval using acquirer price reaction. The RD design essentially compares acquirer price reaction to deals where acquirers intend to issue either above or below the 20% threshold by a small margin, which leads to a discrete change in the requirement of shareholder approval as imposed by the three major exchanges. The key assumption of a valid RD design is that agents cannot precisely manipulate the “running variable.” In our setting, the running variable is the percent of new shares an acquirer intends to issue to finance a deal. If acquirer management—even while having some influence—is unable to precisely manipulate the running variable, a consequence of this is that the variation in treatment near the 20% threshold is randomized as though from a randomized experiment. As such, the RD design provides a clean causal estimate of the effect of shareholder approval on deal quality.
We argue that the key assumption of a valid RD design is satisfied in all-stock deals (i.e., the entire purchase price is paid in stock) used in our analysis. It is true that acquirer management has some control over methods of payment—all-stock, all-cash, or a combination of stock and cash payment—and in the last case, over the faction of payment in stock. However, it is highly unlikely, if not impossible, that in all-stock deals, acquirer management could have precise control over the running variable to avoid the requirement of shareholder approval due to a number of exchange rules and (unforeseen) circumstances associated with M&As that we discuss in our paper. We show that the density function of the running variable in all-stock deals reveals no evidence of precise manipulation by acquirer management around the 20% threshold. A formal McCrary test for smoothness of the running variable further confirms the validity of our RD design.
Using a hand-collected sample of U.S. M&A deals that involve all-stock payment over the period 1995-2015, we find a large and significant 4.3% jump in acquirer announcement returns at the 20% threshold. Given that the average acquirer in the sample used in the RD analysis has a market capitalization of $3.3 billion, a 4.3% jump in stock price around the merger agreement announcement corresponds to value creation of $140 million for acquirer shareholders, suggesting an economically significant value effect. We further show that this positive value effect is concentrated among acquirers with more effective shareholder monitoring as proxied by high institutional ownership, particularly high quasi-indexer ownership, and among acquirers buying target firms with more severe information problems such as unlisted targets or targets with low analyst coverage. We provide some suggestive evidence on the underlying economic mechanisms behind this positive value effect. We find that the requirement of shareholder approval commits acquirer management to seek deals with greater synergies, strengthens its bargaining position vis-à-vis target management, and prevents overpayment. Finally, we show that shareholder approval leads to better post-merger operating performance in acquirers with high institutional (quasi-indexer) ownership.
Our findings have important implications for securities regulators, stock exchanges, and investing public. In November 2015, the NASDAQ requested comments on the 20% rule, specifically whether it is too restrictive and whether the percentage should be higher (i.e., 25%). Institutional investors such as the California Public Employees’ Retirement System—the largest public pension fund in U.S. are in firm support of status quo and argue that any weakening of the NASDAQ’s 20% rule is inconsistent with its goal of preserving and strengthening the quality of its market to protect investors. Our findings in this paper suggest that this listing requirement in the U.S. indeed achieves its intended effect—it empowers shareholders and leads to value-enhancing corporate decisions.
The complete paper is available for download here.
