The Lifecycle of Takeover Defenses
William C. JohnsonJonathan M. Karpoff Sangho Yi
Sawyer Business SchoolFoster School of BusinessSogang Business School
Suffolk UniversityUniversity of WashingtonSogang University
Boston, MA 02108Seattle, WA 98195 Seoul, South Korea
This version: December 8, 2015
Preliminary – Please do not circulate without author consent.
Abstract:
Previous studies on the economic consequences of takeover defenses report inconsistent results. For instance, Gompers, Ishii, and Metrick (2003) and Bebchuk, Cohen, and Ferrell (2009) report that takeover defenses destroy firm value while Smith (2013) and Johnson, Karpoff, and Yi (2015) report exactly the opposite results. Furthermore, previous studies fail to explain why takeover defenses are a prevalent corporate phenomenon if they are value-destroying. We develop a lifecycle view of takeover defenses and use this new paradigm to resolve the inherent conflict within the extant literature. Our lifecycle view of takeover defenses suggests that at the early stage of a firm’s life the bonding benefit of takeover defenses as suggested in Johnson, Karpoff, and Yi (2015) dominate the costs. But as the firm matures, the managerial moral hazard costs increase and the average takeover defense becomes value-decreasing. Our empirical results support a life cycle view of takeover defenses,suggesting that contradicting results within the prior literature are the result of researchers not considering the life cycle effect of takeover defenses. Our life cycle view of takeover defenses also explainswhy a value-maximizing firm may adopt takeover defenses early in its life while takeover defenses are value destroying for large mature firms.
JEL classification: G34, K22, L14
Keywords: Anti-takeover provisions, takeover defenses, IPO, bonding
I. Introduction
Takeover defenses are a prevalent corporate phenomenon and the economic consequences of takeover defenses have been thoroughly examined in the extant literature. However, in spite of numerous previous studies which are focused on the economic consequences of takeover defenses and ample evidence showing the relationship between takeover defenses and firm value and various mechanisms intermediating between them, the current state of the literature is problematic for two reasons.
First, previous studies report inconsistent results. For instance, Gompers, Ishii and Metrick (2003), Masulis, Wang, and Xie (2009) and Bebchuk, Cohen and Ferrell (2009) report that takeover defenses are negatively associated with firm value. These results support the classical agency view of takeover defenses which suggests that takeover defenses entrench managers and allow them to seek their private benefits at the expense of firm value (Easterbrook and Fischel, 1991). On the other hand, contrary to the prediction of the classical agency view of takeover defenses, Field and Karpoff (2002) and Johnson, Karpoff and Yi (2015) show that IPO firms in which managerial moral hazard problem is negligible due to highly level of managerial ownership adopt a significant number of takeover defenses. Furthermore, Field and Karpoff (2002) show that IPO firms with a larger number of takeover defenses tend to show better ex post operating performance. In addition, Johnson, Karpoff and Yi (2015) and Smith (2013) show that the number of takeover defenses is positively associated with the firm value at the IPO stage. Thus, it is clear that there is a contradiction in the literature regarding whether takeover defenses increase or decrease firm value.
Second, the extant literature fails to answer the very basic question of why firms adopt takeover defenses. Gompers, Ishii and Metrick (2003), Masulis, Wang, and Xie (2009) and Bebchuk, Cohen and Ferrell (2009) suggests that takeover defenses aggravate managerial moral hazard problem and destroy firm value. If takeover defenses are value destroying, why would value-maximizing firms adopt takeover defenses which entrench managers thus destroying firm value?[1]
In this paper, we aim at reconciling these conflicting empirical results by developing and investigating a life cycle view of takeover defenses. For the first time in the literature, we examine the evolution of takeover defenses over a firm’s life cycle and the life cycle dynamics of the benefits and costs of takeover defenses. The previous literature examining the economic consequences of corporate takeover defenses employs a cross-sectional approach which eliminates the time dimension from empirical analyses.[2] We argue that the evolution of takeover defenses over a firm’s life cycle underlies the seeming inconsistencies documented in the extant literature. Our life cycle view of takeover defenses is as follows: The costs of corporate takeover defenses start at a negligible level at the early stage of a firm’s life cycle and increase as a firm grows. On the other hand, the benefits of corporate takeover defenses start large at the early stages of a firm’s life cycle and decline as the firm grows. Thus, the net benefit of takeover defenses declines over a firm’s life. Our life cycle view of takeover defenses predicts that young firms have a positive relationship between firm value and takeover defenses but the sign of this relationship reverses and there is a negative relationship between firm value and takeover defenses for mature firms, which we call a value reversal. In our view, the conflicting results reported by previous studies is not a puzzle but only highlights the importance of incorporating the dynamics of takeover defenses over a firm’s life cycle into the empirical analysis.
Our life cycle view of takeover defenses is hinted at in some previous studies. The classical agency view suggests that costs of corporate takeover defenses are smaller at the earlier stage of a firm’s life but grow steadily over a firm’s life cycle as managerial ownership diffuses and founders are likely to leave the firm. In addition, as firms get larger, managers have greater incentives to extract higher rents by their opportunistic behavior.On the other hand, Johnson, Karpoff and Yi (2015) note that the benefit of corporate takeover defenses is biggest at the early stage of a firm’s life cycle and declines as a firm grows. At the early stage of a firm’s life, the firm usually needs stakeholder investment in highly specific assets, which makes the firm’s opportunistic behavior to extract quasi-rents more likely. Furthermore, a young firm does not have a long history over which to build a reputation to honor stakeholders’ implicit claims. This increases the value of corporate takeover defenses as a commitment device to honor these stakeholders’ implicit claims. Furthermore, a young firm typically lacks the resources to construct diffuse distribution channels and instead relies heavily on a few customers and suppliers, which gives large customers and suppliers significant bargaining power over a young firm.[3]
Aside from the implications of the previous studies, our investigation of a life cycle view of takeover defenses is also motivated by anecdotal evidence which suggests that when a firm adopts takeover defenses it considers its current lifecycle stage. For instance, the prospectus[4] of Advantage Learning System states that:
“The provisions of the Company's Articles of Incorporation and By-Laws and the WBCL described in this section may delay or make more difficult acquisitions or changes of control of the Company not approved by the Company's Board of Directors. Such provisions have been implemented to enable the Company, particularly (but not exclusively) in the initial years of its existence as a publicly-traded company, to develop its business in a manner which will foster its long-term growth without disruption caused by the threat of a takeover not deemed by its Board of Directors to be in the best interests of the Company and its shareholders.”
If takeover defenses enhance firm value early in the life cycle of the firm and destroy value later, then a value-maximizing firm would adopt more takeover defenses at the earlier stage of its life when the net benefit of takeover defenses is greater and remove them drastically at the later stage of its life when net benefit of takeover defenses becomes negative. But, this is not the case. On the contrary, a firm tends to adopt a menu of takeover defenses at the IPO stage and seldom revise them as it matures (Coates, 2001). We confirm this stagnation of corporate takeover defenses throughout our empirical results. There are many potential reasons why corporate takeover defenses are insensitive to the changes in various factors which affect the benefits and costs of takeover defenses. First, managers may prohibit firms from removing takeover defenses to protect themselves from disciplinary outside takeover attempts. Second, as Coates (2001) suggests, the power of managers who try to retain takeover defenses to entrench themselves is exactly offset by the power of institutional shareholders which try to remove value-destroying takeover defenses. Third, according to Johnson, Karpoff and Yi (2015), corporate takeover defenses are used as a contractual commitment device to persuade stakeholders to make valuable relationship-specific investments. The irreversible nature of such a comment device may lead to a difficulty in modifying corporate takeover defenses at a later stage of a firm’s life. This suggests that rather than takeover defenses being dynamically modified to maximize firm value at all times, the level of takeover defenses may stagnate, resulting in a value-destroying level of takeover defenses being in force as the firm matures.
The declining net benefit of corporate takeover defenses over a firm’s life cycle and the stagnation of these defenses over a firm’s life cycle are two main themes of our paper. These yield a rich set of testable implications which have not been examined in the extant literature. First, the declining net benefit of corporate takeover defenses over a firm’s life cycle implies that there is a distinct life cycle effect for thevalue impact of takeover defenses. Young firms have a positive relationship between firm value and takeover defenses but the sign of this relationship reverses and there is a negative relationship between firm value and takeover defenses for mature firms. Second, a firm’s needs for making a commitment to honor stakeholders’ claims has a first order effect on the valuation effect of takeover defenses at the earlier stage of its life and managerial moral hazard problem has the first order effect on the valuation effect of takeover defenses when the firm matures. Third, takeover defenses which have a commitment nature will especially enhance the life cycle effect of takeover defenses regarding firm value creation. Takeover defenses used as an irreversible commitment device may lead to ex ante value creation but ex post value destruction and thus cause a value reversal (Stout (2002)). We devise a takeover defense index which we call c-index(commitment index) which is defined as the number of takeover defenses whose sole purpose is to make it hard to remove takeover defenses. This c-index includes supermajority requirement to amend the charter and supermajority to amend the firm bylaws.[5] Some features of c-index are noteworthy. First, the results reported in the Appendix table 6 show that c-index is significantly negatively related to the likelihood of removal of other takeover defenses, which validate our use of c-index to measure the value of commitment in takeover defenses. Second, our use of c-index to measure the value of commitment in takeover defenses is justified by another observation. Takeover defenses which are counted in c-index are adopted at the IPO stage and remain stagnant over a firm’s life, which is consistent with the nature of commitment. On the other hand, other takeover defenses are more flexibly revised at firm discretion over a firm’s life. Thus, our c-index is a useful measure of the degree of stagnation of takeover defenses over time.
Our life cycle of takeover defenses hypothesis explains why previous studies which utilize a cross-sectional approach find inconsistent results concerning the valuation effects of takeover defenses. These studies simply examine the valuation effects of takeover defenses at different stage of a firm’s life. Furthermore, the life cycle of takeover defenses explains the puzzle of why many firms adopt takeover defenses even when such an adoption appears to be value-destroying. When takeover defenses are adopted at the earlier stage of a firm’s life, the benefit of takeover defenses exceeds the cost of takeover defenses. Later, the valuation effects of takeover defenses are reversed as a firm matures but takeover defenses are difficult to remove since takeover defenses are stagnant over time.
Our empirical findings are consistent with our life cycle view of takeover defenses hypothesis. We find that there are two categories of takeover defenses. There are takeover defenses which are stable over time and takeover defenses which are flexibly adopted and removed depending on firm discretion of takeover defense adoption over a firm’s life cycle. The first one corresponds to the bonding hypothesis of takeover defenses suggested by Johnson, Karpoff and Yi (2015) which emphasizes the role of takeover defenses as a commitment device at the earlier stage of a firm’s life. The latter corresponds to the classical agency view of takeover defenses which argues that managers try to entrench themselves by adopting takeover defenses. Thus, we find that certain takeover defenses are adopted at different stages in a firm’s life for different purposes.Previous studies which focus on changes in aggregate score takeover defense index (Gompers, Ishii, and Metrick (2003)) cannot tease out the importance of individual takeover defenses. Furthermore, we find that this distinction between two categories of takeover defenses plays a significant role in the life cycle effect of takeover defenses and value reversal. We introduce the commitment index (c-index) which includes firm characteristics making it hard to revisefirm takeover defenses. Our empirical analyses confirms that it is the c-index that causes a reversal in the sign of the relationship between firm value and takeover defenses. This is consistent with Stout (2002) who argues that takeover defenses used as a commitment to honor stakeholder claims will create ex ante efficiency and ex post inefficiency. Also, consistent with the bonding hypothesis of takeover defenses (Johnson, Kaporff and Yi (2015), we find that takeover defenses not in the c-index (non-c-index) do not show value reversal in the relationship between firm value and takeover defenses but only decreases firm value.
In addition, we find evidence of a life cycle effect of takeover defenses on firm value. In the year of an IPO and in the year after the IPO, we find that the industry adjusted Tobin’s Q of firms which have above the median e-index is greater than the industry adjusted Tobin’s Q of firms which have below the median e-index. However, starting 3-4 after the IPO event, we find that firms with more than the sample median number of takeover defenses have a significantly lower industry adjusted Tobin’s Q compared with the firms with fewer than the sample median number of takeover defenses. This indicates that there is a reversal in the valuation effects of takeover defenses over a firm’s life cycle. We find a similar pattern of reversal of the valuation effects of takeover defenses when we use classified board and c-index instead of e-index. Consistent with the aforementioned univariate results, in multivariate analysis using the exact control variables utilized by Gompers, Ishii and Metrick (2003), we find that the interaction between e-index and firms being more than four years from their IPO is significantly negatively related to industry adjusted Tobin’s Q at the 1% level. This result still holds when we utilize staggered board or c-index as the measure for takeover defenses.
We also find evidence that takeover defenses which are utilized as a commitment device and are hard to revisearethe driving force which leads to the value reversal we document. We find that the valuation effect of c-index goes from value-increasing to value-decreasing over a firm’s life cycle. On the contrary, other takeover defenses which can be easily removed by the board without shareholder approval does not show a strong value reversal. We find that the valuation effect of e-index goes from value-increasing to value-decreasing over a firm’s life cycle even when we utilize firm age and sales growth rate of the firm or the firm’s industry as alternative proxies of IPO firm maturity.
Importantly, we find that a firm’s needs to promote stakeholder relationships using takeover defenses as a commitment device make takeover defenses value-increasing but these takeover defenses become value-destroying since takeover defenses are insensitive to changes in the economic environment which affect the valuation effect of takeover defenses. We use two variables to capture a firm’s needs to promote stakeholder relationships using takeover defenses as a commitment device. One is an indicator variable for having a large customer and the other is and indicator for having strategic alliances. The fact that a firm relies on a large customer indicates that the firm fails to diversify its stakeholder-base possibly due to undeveloped distribution channels and a small number of large customers will hold a significant bargaining power over the firm (Johnson, Kang, Masulis and Yi, 2015). Furthermore, previous studies suggest that sustaining strategic alliances require underlying relationship-specific investments by strategic alliance partners. Thus, these two variables can effectively capture a firm’s needs to promote stakeholder relationships using takeover defenses as a commitment device. We find that when the firm is older than the median firm the interaction between e-index, and the large customer indicator variable is significantly negatively related to industry-adjusted Tobin’s Q. We find similar results when we use strategic alliance measure capture a firm’s needs to promote stakeholder relationships. As the firm matures, the importance of the firm’s relationships with outside parties such as customers and alliance partners declines, making the commitment value of takeover defenses lower as time passes.