Manager replacement, employee protest, and corporate control

Some motivating examples of the assumptions of the model, such as the non-contractibility nature of the employee–manager relationship, are discussed in Sect. 2.

When a talented employee has received the firm’s specific training, she is an intangible asset to the firm. Therefore, her threat to leave means that the company may need to replace her with a less capable employee or with a non-trained worker, which would be a loss for the company.

There are many other papers that discuss internal corporate governance mechanism focusing on ownership structure, board composition, board independency, board size, managers’ compensation package, and internal auditing system. See, for example, the papers by Gompers et al. (2003), Landier et al. (2009), Boone et al. (2007), Bebchuk et al. (2009), Frantz and Instefjord (2009), Edwards et al. (2009).

See EXPANSIÓN, December 11, 2007, in the article “Empleados brillantes... y muy fieles” (“Clever employees... and very faithful”).

See, for example, Armstrom and Taylor (2014). In their book “A Handbook of Human Resource Management Practices”, Armstrom and Taylor indicate that “...how people are treated in such areas as recruitment, performance reviews, promotion, career development, reward, involvement and participation.... will be particularly important (to human resource management)”.

http://​www.​veooz.​com/​news/​6HHYALR.​html. This news report states that not only employees supported Mr. Demoulas. Also customers and suppliers may. A grocery buyer of Marker Basket, Brian McGuire, said “He’s one of the people. He’s one of us”. “He knows all his employees names. He comes in and talks to the customers”.

http://​www.​itpro.​co.​uk/​strategy/​22367/​facebook-ceo-to-donate-120m-to-san-francisco-schools.

http://​www.​fin24.​com/​Economy/​CEOs-donate-millions-to-charity-20110906.

One example of the employee influential power at the firm can be the number of seats held by employees on the board. More seats held by employees mean more employee influential power at the firm. Another example is the employee’s bargaining power. If an employee has some special skills and would be difficult to replace, she would have more influential power in the firm (Landier et al. 2009; Marino et al. 2010). An extreme case of employee influential power at a firm is when employees have ownership of the firm. This is not uncommon in the auditing, consulting, and private equity industries. In the model, the relative influential power is an endogenous decision made by the shareholder. However, the equilibrium results remain similar if s is exogenous. The motivation to endogenize the decision is to check whether the shareholder is willing to give up some power to the employees in the setup.

The employee contract is often decided by the manager. Here I assume that the shareholder is in charge of all contracts because contracts are verifiable and can therefore be controlled by the shareholder. Even if the contract is decided by the manager, the final wage payments are usually paid by the shareholder in a firm. Therefore, unless the manager can bribe the employee (which is an interesting case but is not studied in this paper) the final results remain the same.

The parameter r is assumed to be small enough to ensure \(q^m\in \left[ 0,1\right] \). A more generalized probability function such as \(\beta a_m+\varphi e+\theta a_m e\) also leads to a similar result.

The continuous investment strategy leads to the same static comparative analysis, but the complexity of the functional form does not allow me to find a closed-form solution. However, some numerical example analysis also shows similar results to those of the discrete investment case. The analysis is available upon request.

Building a good relationship usually takes a long time. Therefore, in this static setup, the investment from the outside manager is assumed to be 0. However, this assumption is not critical to the results. Suppose a dynamic setting with two periods when the outside manager can make a similar investment in the second period. Then, the employee has to decide whether to support the incumbent manager or to have a new manager. The decision depends on the employee’s discount rate (because if the outside manager invests in a good relationship with the employee, it is in the second period, and its effect on the employee’s utility is discounted) and the probability that the outside manager will invest in the relationship. Once the outside manager is hired, his incentive to invest is very low. The probability of making an investment in the relationship decreases as the replacement threat decreases. Therefore, this assumption is not critical for the main results. Moreover, the relationship is private/personal. It is logical to assume \(I_{OM}=0\).

In this paper, I focus on a specific relationship between the incumbent manager and the employee. Once the incumbent manager is fired, the employee can no longer enjoy the benefits from the close relationship. For instance, donating to local environment committee, to local schools or to employee–friendly politicians builds up a relation between the incumbent manager and local government or politicians. Employees may feel better to work with the incumbent manager because of his contribution to the local environment and education systems or may benefit from employee–friendly policies. When the donations stop, employees’ proudness disappears or a new politician comes into the office and implements non-employee–friendly policies. Hence, such manager-employee relationship is private to the incumbent manager.

The replacement probability \(\pi \) depends on the structure of the outside manager’s market. For example, if the market is very competitive (i.e., there are a lot of capable managers and the manager’s ability is easy to learn), it is easier for the shareholder to find a better manager from the outside market and \(\pi \) is close to 1. Otherwise, \(\pi \) is low and it is harder for the shareholder to find a better alternative.

In a more general setup, the probability \(\pi \) may be a decision variable for the shareholder and may depend on the comparison between the advantages of having a more capable manager and disadvantages such as the cost of searching for a new manager or the cost of firing the incumbent manager.

In the model, I do not consider the case in which the employee can also protest against a bad manager. This is because if the employee wants a bad incumbent manager to be replaced, she just needs to keep silent.

The cost can come from the negotiating effort, losing the current job if she threatens to leave, and so on.

I ignore the moral hazard problem between the shareholder and the manager in the paper to simplify the model. The main conclusions made in the paper are not driven by this assumption. In the model, the main decision made by the incumbent manager is the relationship investment \(I_{IM}\), which is not contractable. Then, the manager’s wage contract is not influenced by the investment decision if the employee’s effort is discrete. We can think of \(a_m\) as the optimal working effort in a moral hazard setup.

The parameter \(\gamma \) represents those benefits other than wages enjoyed by managers. Examples for \(\gamma \) might be having a well-equipped office, good working welfare, controlling power or even the pride of being a manager, etc.

The private benefits for the outside manager is normalized to 0 for the simplicity of notations. This normalization does not change the equilibrium results. Suppose that the outside manager can also enjoy private benefits if he becomes the new manager of the firm (\(\gamma _{OM}>0\)). His decision on investing in a good manager-employee relationship depends not only on private benefits, but also on the outside labor market (\(\pi \)). The current equilibrium results still hold if the incumbent manager has a higher incentive than the outside manager to invest in a good relationship with the employee.

The effect of a good relationship on the employee can be modeled in other ways. For example, a good relationship increases the project’s probability of success. Either modeling method leads to the same results.

If there are fixed working costs, the function \(g(I_m)\) can also be understood as a reduction in the employee’s fixed cost. Suppose that the cost function is \(T(f)=G-g(I_m)+C_m(f)\), where \(G-g(I_m)\) is the fixed cost while the second part is the variable cost. Thus, \(g(I_m)\) captures the reduction in the fixed cost, due to the existence of the manager’s investment.

I concentrate on this particular situation because the results of other situations are similar to this because they can simply be interpreted as a redefinition of \(\gamma \). Moreover, \(\underline{\omega }_{LL}\) can be smaller or equal to \(\underline{U}\). Here they are assumed to be equal (without changing the main results in the paper), just to simplify notations and functional form in the analysis of equilibrium behavior. However, it is more interesting to analyze the case of \(\underline{U}-\gamma < \underline{\omega }_{LL} \). When \(\underline{U}-\gamma \le \underline{\omega }_{LL} \), the incumbent manager gains less than the reservation utility even with the private benefit as a manager in the firm. There would be no incentive for the incumbent manager to stay in the firm and invest in a good manager-employee relationship. The proof of Lemma 2 in the “Appendix” contains a detailed discussion and calculation of other situations.

Based on Assumption 1, \(EV^m_{SH} (e=1)>EV^m_{SH}(e=0)\), irrespective of whether or not there is employee–manager relationship investment. This means that \(rR-1>0\) always holds, which leads to Eq. (11).

The incumbent manager’s incentive to invest in the employee–manager relationship is increasing in his own ability and decreasing in the ability of the outside manager. This, together with \(\Phi _{I^*}>0\), implies that

In the model, the power-sharing strategy is endogenously determined by the shareholder because it is interesting to check whether the shareholder is willing to share power with the employee when there is good manager-employee relationship. However, it can be exogenous without changing the current results.

This statement holds if

This, together with \(\Phi _{I^*}>0\) (Case 1), implies that the statement holds if

Please refer to Friebel and Raith (2004) for the conflicts among managers.