اثرات تهدیدهای تصاحب در سهامداران و ارزش شرکت
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|23110||2006||42 صفحه PDF||سفارش دهید||18310 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Economic Behavior & Organization, Volume 59, Issue 1, January 2006, Pages 45–68
We study the role of takeover threats as a corporate control mechanism using [Aghion, P., Tirole, J., 1997. Formal and real authority in organizations. Journal of Political Economy 105, 1–29] model of formal and real authority. Shareholders do not monitor the manager's actions since ownership is widely dispersed. A corporate raider may monitor and step in if a profit opportunity exists. In our model, a takeover threat decreases managerial effort and may increase the value of the shares. We also look at the extent of anti-takeover defenses favored by individual shareholders and institutional investors.
In many corporations the owners are rarely those who control the day-to-day operation. This separation of ownership and control can give rise to agency problems. A manager may have an incentive to pursue his own interests at the expense of the owners. The owners can alleviate this problem by monitoring his actions. However, when ownership is widely dispersed, and monitoring is costly, no shareholder has an incentive to do so. Alternatively, an outside raider can monitor the actions of the manager. If she2 feels that he does not maximize the value of the firm, she can choose to take control and overrule the manager. In this paper, we study the effects of such a takeover threat on the actions of the manager, the value of the shares, and the value of the firm. In the agency literature, it is usually assumed that if a principal increases her monitoring effort, the agent will increase his effort as well.3 Stronger monitoring increases the probability of being caught shirking, hence the incentive to shirk decreases. In a recent paper, however, Aghion and Tirole (1997) argue that the opposite may also be true. Suppose that an agent can take initiatives and derive private benefits from doing so. Then, stronger monitoring implies a higher probability that the agent is overruled by the principal lowering the agent's incentive to take initiatives, which may hurt the principal as well. In that framework, effort levels of the principal and agent are strategic substitutes,4 rather than strategic complements as they are in the standard agency literature. This set-up seems especially relevant to analyze the relationship between managers and shareholders. Managers are usually expected to take initiatives and look for new opportunities, rather than being passive and perform some pre-defined tasks. Managers also obtain some private benefits from taking these initiatives, such as status, prestige, perquisites, satisfaction from performing certain tasks, and possible side-payments. Giving managers the freedom to take initiatives may be desirable if the projects that they prefer also yield higher profits to the firm's shareholders. However, a complication may arise when the interests of managers are not fully aligned with the interests of shareholders. In that case, shareholders will impose a certain degree of control upon managers. Using the framework of Aghion and Tirole, Burkart et al. (1997) study the role of direct shareholder monitoring, and show that for a large shareholder, the incentive to monitor is increasing in α, the fraction of shares she owns. However, there is a trade-off between costs and benefits of monitoring. Too much monitoring will destroy managerial initiatives. Too little monitoring will give too much discretionary power to managers. Therefore, the optimal ownership structure is characterized by some intermediate value of α. The authors thus show that α can be used as a commitment device. Burkart et al. only study one particular corporate control mechanism. Their model requires that there is a large shareholder who monitors the manager. Therefore, the model is appropriate for analyzing a corporate governance system that relies on the role of monitoring by a large shareholder. Such a system exists in Continental Europe and Japan. In Anglo-Saxon countries, however, the corporate governance structure is generally believed to be characterized by the presence of many small shareholders.5 In such a system the role of direct shareholder monitoring is much less significant. In the absence of direct monitoring, takeover threats are often used as a corporate control mechanism (see also Hart, 1999). In this paper we use the Aghion and Tirole framework to study the role of takeover threats as a corporate control mechanism. Our analysis is complementary to that of Burkart et al. We assume that there are many small shareholders who do not directly monitor the manager because of a free-rider problem. The manager has full discretionary power to decide which project should be undertaken. The choice of project depends on the size of private benefits that he obtains. Following Aghion and Tirole, we assume that the manager's most preferred project also yields positive profits for shareholders, yet a different project exists that yields higher shareholder value. A raider may investigate the possibility of a takeover. She will step in if there is a profit opportunity, which is the case if the project chosen by the manager diverges sufficiently from the project that is optimal from the point of view of the shareholders. If the raider takes over the firm, she pays some fixed premium for the shares and implements her optimal project. In this paper, we study the following issues. First, we show that due to a takeover threat, the effort exerted by the manager will decrease. It is usually argued that a takeover threat will discipline the manager and increase his effort (see for example Scharfstein, 1988), yet in our model, a takeover threat introduces the possibility that the manager will be overruled. If he is overruled, the effort he has taken in evaluating different projects will be wasted. Therefore, a takeover threat reduces the incentive to make such efforts in the first place. This effect is similar to that described in Kahn and Huberman (1988). In their model, employees who face the possibility of being fired have less incentive to invest in firm-specific human capital. In Shleifer and Summers (1988), takeovers are a mechanism used to break the implicit contract between managers and workers. Therefore, workers are less inclined to engage in such contracts in the first place. Due to a takeover threat, the workers may then be reluctant to invest in relationship-specific human capital. In our paper, a takeover threat makes managers more reluctant to invest in relationship-specific information. The mechanism, however, is fundamentally different from Shleifer and Summers. Second, we study how the takeover threat affects the value of the shares. We show that shareholders prefer not to have a takeover threat when the takeover premium is low enough. A takeover threat has two effects. On the one hand, if there is a takeover, shareholders earn more since they receive more for their shares than they would in the absence of a takeover. On the other hand, as we saw above, the effort of the manager is lower with a takeover threat, which reduces shareholder value. In our model, for reasonable values of the takeover premium, the second effect may dominate. Third, we study how a takeover threat affects the value of the firm. We show that this effect is ambiguous. Again, there are two countervailing effects. On the one hand, if there is a takeover, the raider will implement a project that yields higher firm value. On the other hand, the takeover threat reduces the manager's effort. The net effect is ambiguous. The value of the firm decreases with a takeover threat when the interests of shareholders and manager are sufficiently aligned and when the private benefits of the manager are sufficiently high that he has a strong incentive to exert effort. Fourth, we study the extent of anti-takeover defences that would be preferred by shareholders. To do so, we decompose shareholders into individual shareholders who have shares only in the target firm and institutional shareholders who also have shares in the raider. We assume that individual shareholders are interested in maximizing the value of the shares of the target firm whereas institutional shareholders are interested in maximizing the value of the firm: they will ultimately receive this value, either through the shares of the target firm, or through the shares of the raider.6,7 We show that in cases where managerial interests are relatively strongly aligned with shareholders’ interests, the preferences of these parties coincide. They both want a value for the takeover premium that is so high that there is no effective takeover threat. In this case, a raider would not be able to add much value to the firm, and a takeover threat mainly serves to reduce managerial effort, which is bad news for all shareholders. When managerial interests are weakly aligned with shareholders’ interests, this is no longer true. Institutional shareholders now want anti-takeover defences to be as low as possible, as a takeover threat now adds value to the firm. Individual shareholders prefer some intermediate value, balancing the trade-off between managerial effort and the premium they receive should a takeover occur. This paper is organized as follows. In Section 2, we present the monitoring technology used in this paper, which is based on Aghion and Tirole. Section 3 studies the benchmark model with dispersed ownership and no takeover threat, and in Section 4 we study the case where a takeover threat is present. Section 5 derives the necessary conditions for our analysis, which allows us to describe the full model in Section 6. The properties of the model are analyzed in Section 7. Comparative statics are studied in Section 8. Section 9 concludes.
نتیجه گیری انگلیسی
This paper has studied the effect of a takeover threat as a corporate control mechanism, using Aghion and Tirole's model of formal and real authority. Our paper is complementary to Burkart et al. who use the same framework to study the effect of large shareholder monitoring. As our first result, we found that due to a takeover threat, the effort exerted by the manager will decrease. Thus, whereas it is usually assumed that a takeover threat disciplines managers, in our model the opposite is true. Here, the manager has less incentive to exert effort in trying to learn the profitable opportunities of the firm when he knows that he can be overruled.20 Our further results hinge on the extent to which the interests of shareholders are aligned with that of the manager. Suppose that these interests are closely aligned in the sense that projects yielding the highest private benefits to the manager also yield a relatively high return to shareholders, and moreover the manager's absolute private benefits are high such that he has a strong incentive to find the best project. In such a case, a takeover threat will only destroy value. It lowers managerial effort, whereas the benefits in terms of a potentially higher return are only modest. A takeover threat decreases the value of the shares and the value of the firm. In this case, individual shareholders and institutional shareholders21 prefer a prohibitively high takeover premium such that any raider is deterred from monitoring the firm and attempting a takeover. Now consider the case in which interests of shareholders are only weakly aligned with that of the manager, in the sense that projects yielding the highest private benefits to the manager only yield a relatively low return to shareholders, and moreover the manager's private benefits are relatively low such that he does not have a strong incentive to find the best project. In such a case, the value of the firm increases with a takeover threat. If the takeover is successful, the project that will be implemented by the raider is much more valuable for the firm than the project that would have been implemented by the initial manager. This benefit outweighs the decrease in managerial effort brought about by the takeover threat. In case of a takeover, however, individual shareholders will not receive the full value of the firm, they only receive some takeover premium beyond what the value of the shares would have been without a takeover threat. They thus face a trade-off. A higher takeover premium makes a takeover less likely, but it increases the value of the shares by a higher amount should a takeover occur. We find that in this case, a takeover threat makes individual shareholders better off if the takeover premium is relatively high, but it makes them worse off if the takeover premium is relatively low. For the preferred value of the takeover premium (or, more generally, the preferred extent of anti-takeover defences), this implies the following. Institutional shareholders will oppose any anti-takeover defences and prefer no takeover premium. Individual shareholders prefer an intermediate value of the takeover premium that will not deter a raider from monitoring but still allows them to obtain substantial gains should a potential raider attempt a takeover. In the real world, one often observes that the value of the shares increases when a takeover is announced. That observation is consistent with what our model assumes. When a raider has decided to take over a firm, he pays a premium ρ over the original value of the shares, which implies that the price at which the shares are traded on the stock market also increases by this amount. 22 Hence, an actual takeover always makes shareholders better off. What we have shown in this paper, however, is that the mere threat of a takeover could make them worse off. 23 Because of the ambiguous results we find, it is hard to test our model empirically, yet this also implies that our results may explain the conflicting empirical evidence with respect to, for example, the effect of the adoption of anti-takeover defences on share prices.24 Following Theorem 4, with View the MathML sourceρ<ρinc* or λb > 1/2, the adoption of anti-takeover provisions leads to an increase in share prices. This is what Linn and McConnell (1983) find. Ryngaert (1988) draws a similar conclusion. Comment and Schwert, in a very comprehensive survey article on this issue, conclude that the adoption of poison pills and other anti-takeover provisions have lead to higher takeover premiums but have not systematically deterred takeovers. This suggests that such provisions are good for shareholders. Mahoney and Mahoney (1993) find no significant effect in the period 1974–1979, but a positive effect in the period 1980–1988. However, McWilliams and Sen (1997) find a negative effect of anti-takeover provisions on share prices. This is consistent with our model for the case of View the MathML sourceρ>ρinc* and λb < 1/2. One intriguing implication of our model is that different shareholders in the target firm may have different preferences with respect to the extent of anti-takeover defences the firm installs. These preferences are dependent on whether or not shareholders also hold shares in the raiding firm. Suppose that institutional shareholders hold shares in the raiding firm, but individual shareholders do not. Our results then suggest that individual shareholders may lobby for high anti-takeover defences whereas institutional shareholders want no such defences. There is an empirical literature that addresses such differences. Jarrell and Poulsen (1987) find a relation between the share price reaction to the announcement of anti-takeover defences, and the level of institutional shareholdings. Agrawal and Mandelker (1990) also find such a relation. Brickley et al. (1988) study the voting on anti-takeover amendments of different types of investors. However, these papers focus only on the different incentives to monitor by the different types of investors. They do not look at the effect of having shares in the raiding firm on the preferred extent of anti-takeover defences that we find in our model. In this paper, we made a host of simplifying assumptions. Obviously, our portrayal of the way firms are run and how takeovers occur in our model is overly simplistic. For example, takeovers that occur because of possible synergies are not taken into account. Nevertheless, we feel that our model yields a host of interesting insights that would also apply to richer models of takeovers. Other simplifying assumptions include the following. We assumed that the manager does not hold shares and that his remuneration is in no way related to the ultimate value of the firm. Burkart et al. also make this assumption and show that relaxing it does not qualitatively change their results. We also assumed that there is only one possible raider. It is often argued that when there is a takeover attempt, rival bidders may appear. In our paper, this is not a problem. In our model, a bidder attempting a takeover has private information with respect to the project that is best for the firm; therefore, it is not likely that a bidding war will ensue. A second raider, observing the takeover attempt of the first one, first has to invest and be successful in information acquisition. If she does learn the payoffs of all projects and decides to engage in a bidding war, the shares will be bid up to Π, and takeover profits are zero. Hence, a priori, a rival bidder has no incentive to exert effort. 25