1
The Macrotheme Review
A multidisciplinary journal of global macro trends
Board Structure Tradeoffs: New Evidence Using Multi-objective Genetic Algorithm
Aymen AMMARI, Abderrazak ELLOUZE
QUARG, THEMA, High Business School of Tunis, University of Manouba, Tunisia, Higher School of Digital Economy, University of Manouba, Tunisia.
Abstract
Optimal corporate board structure, with a minimum level of agency costs, is considered to have significant implications for firm performance. When selecting board structure in practice, French firms may be motivated by resource dependency, but long-term business planning and multi-criteria decision-making requires method-based new decision support tools. These tools must be able to predict the change in the board structure storage with sufficient accuracy, and must allow exploring dynamic scenarios such as the economic Darwinism. In this context, (Ning et al., 2010) argue that economic Darwinism would eliminate a large corporate board if a board with a large number of directors is harmful to firm value. Similarly, if large boards were better, economic Darwinism would eliminate small boards. The simulation of natural evolutionary processes of human beings results in stochastic optimization techniques called evolutionary algorithms (EAs) that can often outperform conventional optimization. Among them, genetic algorithms are perhaps the most widely known type of evolutionary algorithms used today. In this paper, a multi-objective genetic algorithm optimization (MOGA) is applied to determine optimal board structure of board size, board independence, number of board’s committees and committees’ independence that simultaneously maximize firm performance and minimize agency cost. The methodology proposed has more Pareto optimal solutions. However, it captures the Pareto front for the dual objective problem indicating a number of useful Pareto optimal designs in order to provide a variety of compromise solutions to help the decision maker.
Keywords: OPTIMAL CORPORATE BOARD, FIRM PERFORMANCE, GENETIC ALGORITHM
1. Introduction
Corporate governance, and in particular the issue of control of corporate boards has received considerable attention recently, in the wake of corporate scandals afflicting the likes of Enron, Tyco… and searching for an optimal board structure is a key concern for researchers into the corporate governance field. In France, board of directors, to apply their monitoring duty with due diligence has been mainly made upon its inclination to implement corporate governance reforms suggested by three important reports: Vienot I (1995), Vienot II (1999), and Bouton (2002). This was an outcome, in great part, of three events. Basically, France perceived an important spread in the degree of foreign institutional ownership, mostly from Anglo-American mutual and pension funds. Managing 42.4 % of the equity capital of CAC 40 firms in 20021, these foreign institutional investors have applied a powerful institutional pressure for the implementation in France of corporate governance restructurings comparable to those which were realized in their countries of origin and this include board structure. In addition, the French corporate governance reports, Vienot I (1995), Vienot II (1999), and Bouton (2002), were powerfully supported by the French business confederation (MEDEF). This robust encouragement has been driven by a worry that the French state installs strict and obligatory laws for controlling the process and the structure of the board of directors after several governance scandals. Third, the implementation of the Vienot I (1995), Vienot II (1999), and Bouton (2002) corporate governance reforms, proposing an optimal board size, the establishing of optimal number of specialized committees as well as an optimal proportion of independent directors, is exclusively voluntary. Similarly, the French financial markets authority (AMF), to adopt practices recommended by French reports,
1 Bank of France Bulletin, 2004
2 does not necessitate listed companies but only recommends corporations to be relevant to one of these reports as a good word for designing the annual report. As the implementation of these reforms is a board concern, it symbolizes a robust indication showing the extent to which board members are motivated to control and monitor management. This is mainly sincere for board members that resist to institutional pressures. In our sample, a large number of firm boards have adopted over the 2001–2013 period the set of reforms promoted by French corporate governance reports such as varying the size of their boards, increasing the ratio of independent director and creating serve committees. On the other hand, due to the discretionary environment of such recommendations, several other corporate boards have resisted such institutional pressures and did not act in accordance with these best practices. This high variance along key variables capturing board structure and the presence of serve committees establishes a relevant empirical context to test board structure evolution.
An additional motivating feature of the French framework is allied to the social structure of elites in France. Without a doubt, there is an extremely dense social network of ties surrounded by the French elite, institutionally formed from beginning to end by the grands corps (“grande écoles”) and business associations, simplifying the transmission of information associated with individuals’ behaviors (Burt et al., 2000; Kadushin, 1995; Maclean et al., 2006). That is, indications related to board members’ behavior within the board and board structure, designed by the total and kind of reforms they implements, are apprehended by the diverse actors in the network. Optimal board structure consecutively, enlarges the observability of behaviors targeting at increasing supervise over management in the labor market for board of directors. Lastly, despite an increasingly growing diffusion of corporate committees and board independence percentage in French and continental European firms, to our best knowledge empirical investigation of the search of optimal number of committees or independence percentage is nonexistent. This literature gap motivated our use of the French context to try to find its optimal corporate board structure. The developments cited in the previous paragraph propose that the main view among regulators is that it is in the interest of shareholders for corporate boards to be configured by an optimal structure. This sight appears to be driven by agency reflections, that is, that only an optimal board structure can effectively curtail agency problems. While agency problems are evidently significant, other concerns may affect the conclusion that boards should be efficient and contribute in firm performance. More specifically, the board’s decisions are based on the information available to its members; information provided by them. From this sight, firms are better off with large and more independent boards. Each new board member brings both expertise and access to resources. Having more board members, more independence and more monitoring committees would, therefore, provide the firm with greater knowledge and access to resources. These resources could include access to markets, access to new and better technologies... Large independent boards are more likely to have directors with greater range in culture and industry experience. This variety permits the board members to offer management with high quality advice (Zahra & Pearce, 1989). However, Lipton &
Lorsch (1992) and Jensen (1993) claim that large boards and adding serve committees can be costly. Larger boards amplify operational complexity and increase the potential for disagreement among members, therefore leading to more director free-riding problems and internal conflicts among directors. The implantation of serve committees and appointing more independent director produces greater costs for the firm and thus affecting negatively firm performance. This leads to some unique trade-offs in corporate board structure which refers to the idea that a company chooses how many board directors, how much independence percentage and how many board committees to use by balancing the costs and benefits. Determining an optimal board structure is a chief requirement of any firm's corporate governance department and requires method-based new decision support tools. These tools must be able to predict the change in the board structure storage with sufficient accuracy, and must allow exploring dynamic scenarios such as the economic Darwinism. In this context, (Ning et al., 2010) claimed that economic Darwinism would eliminate a large corporate board if a board with a large number of directors is harmful to firm value. Likewise, if large boards were better, economic Darwinism would eliminate small boards. Simulating natural evolutionary processes of human beings results in stochastic optimization techniques called evolutionary algorithms (EAs), that can often outperform conventional optimization. Among them, genetic algorithms are perhaps the most widely known type of evolutionary algorithms used today. Genetic algorithms are techniques of inductive learning built on adaptive search practices, which have the power of handling accumulative information on the subject of an unknown exploration space with the objective of forwarding consecutive searches into the most appropriate subspaces, as a simulation of the biological evolution (Vafaie & De Jong, 1992). A candidate solution is distinguished by processes of a linear string correspondingly to a chromosome. A population evolves toward better solutions, and a fitness function put a figure on the appropriateness of each solution. As a result, this practice has been effectively operated to the examination of different scenarios, with the object of determining the most relevant features for the clarification of a certain occurrence (Huang et al., 2007; Rozsypal & Kubat, 2003; Yang & Honavar, 1998). The current advances into the quantitative finance area leads to the emergence of complex mathematical paradigms for the description of different financial occurrences, which are often categorized by a lack of analytic illustration. In this
3 view, a number of metaheuristic algorithms have revealed to be an appropriate approach for giving a lecture to these financial complications. Specifically, genetic algorithms have been effectively used to the examination of different financial circumstances with promising findings. Consequently, genetic algorithms have been newly applied in value-at- risk computing (Sharma et al., 2015), optimal insurance risk allocation (Ha, 2013), bankruptcy research (Davalos et al., 2014; Shin & Lee, 2002; Wu, et al.,2007), exchange rates prediction (Vasilakis et al.,2013), portfolio optimization (Chang et al., 2009; Oh et al., 2005), financial failures forecasting (Chen, 2014), financial fraud detection (Hoogs et al., 2007), and stock markets prediction (Ghoshal et al., 2011; Karimi et al., 2014; Leighet al., 2002). In this paper, a multi-objective genetic algorithm optimization (MOGA) is applied to determine optimal board structure of board size, board independence, number of board’s committees and committees’ independence that simultaneously maximize firm performance and minimize agency cost.
2. Literature review
The empirical results in previous studies are mixed regarding the relationship between firm value and board structure. The finance literature has generally found evidence consistent with the agency theory perspective that a smaller board is related to better firm performance (Yermack, 1996; Gertner & Kaplan, 1996; Eisenberg et al., 1998; Denis & Sarin, 1999). Due to coordination costs and free rider problems inherent in large boards, shareholder groups generally favor smaller boards and have pressured companies to reduce board size (Gertner & Kaplan, 1996).
Some management studies, however, have found a large board to be better (Gales & Kesner, 1994; Dalton et al., 1999), and these studies seem consistent with resource dependency theory, which supports a positive relationship between board size and firm performance (Dalton et al., 1999). The independence of administrator is a critical issue of opportunistic manager behavior. However, the independent administrator role is limited by many specific factors such as information availability and costs related to specific information. (Hadani et al., 2011) indicated that the effectiveness of the board improves when the majority of the directors are independently sat on its committees. However, they also found that this improvement requires a substantial cost. Restoring good governance happens through a strengthening of the control of the administrators on managerial decisions. In this perspective, the committees provide administrators with the means and structure that permits them to exercise effective control over a number of issues deemed crucial to defending the interests of investors (Braiotta and Sommer, 1987)
A. Agency Theory Perception
Lipton & Lorsch (1992) and Jensen (1993) claimed that large board and adding serve committees can be costly.
Larger boards amplify operational complexity and increase the potential for disagreement among members, consequently leading to more director free-riding problems and internal conflicts among directors. This produces greater costs for the firm and affecting negatively firm performance. Empirical investigation has found consistent results concerning board structure and firm value from the agency theory perception. Yermack (1996) uses an illustration of large U.S. publicly traded firms from Forbes 500 and uncovers a consistent and inverse relationship between Tobin’s Q and board size even after controlling for firm size, growth opportunities, market performance, business diversification, board composition, and ownership structure. Eisenberg et al. (1998), using a sample of 879 midsize and small Finnish firms, approved similar conclusions. Both Tufano & Sevick (1997) and Dann et al., (2000) realize that a small board is more likely to be related with a lower expense ratio in the fund industry.
B. Resource Dependency Theory Perception
According to Resource Dependency Theory, firms are better off with large and more independent boards. Each new board member brings both expertise and access to resources. Having more board members, more independence and more monitoring committees would, therefore, provide the firm with greater knowledge and access to resources. These resources could comprise access to markets, access to new and better technologies... Large independent boards are more likely to have directors with greater range in culture and industry experience. This variety permits the board members to offer management with high quality advice (Zahra & Pearce, 1989). There is empirical funding for the resource dependency theory. Dalton et al. (1999) find a significant and positive board size and performance relationship in a sample of 20,620 firms. Booth & Deli (1996) claimed that environmental ambiguity usually leads to large board size.
Probably, a larger board permits firm access to the expertise needed to overcome this uncertainty. Pfeffer (1973) finds that the board size is positively linked with the sources of funding in health care firms. This yet again proposes that larger boards offer greater resources than smaller boards.
C. Darwinism in corporate governance
4 According to Jensen (2001), corporations that try to act only through the prescriptions of stakeholder theory will ultimately be unsuccessful since natural selection will remove them if they are in competition with firms that are behaving so as to maximize value. Charreaux (1987) indicates that the positive theory of agency involves ideological inferences.
Since contractual forms are in competition, and only the most adapted survive, it certainly guides to normative conclusions2. In particular, the assumptions of the positive theory are liberal and suitable with the recent economy of property rights. This flexibility occurs through its aptitude to reduce contract costs (agency costs: surveillance costs, and transaction costs) and cognitive costs associated to behavioral biases. Immediately adapt and survive resumes to manage potential conflicts among stakeholders at minimum cost. The persistence of corporate governance system and each mechanism is related to their ability to reduce these costs and only efficient and optimal governance system can survive.
Adapt and survive is accompanying with efficiency, it allow corporation to hold more power across hostile growth policies and establishing more networks. A firm that wants to survive in a background exposed to the "Darwinian selection" is destined by this means to adapt. "Adapt or die!" is the roadmap to success as well-defined by Heinrich &
Betts (2003), Beerel (2009), Morçöl (2007), Appay (2005) and Nelson (2006). According to Pascale (2001), an all- powerful environment with unpredictable fluctuations can control the firm from top to bottom. There are no refuges from this “Darwinian jungle”, and it is not getting easier. In this condition, the environment necessitates a practice leading to select the most appropriate organizational systems. Hannan & Freeman (1989) argued that firms suffering from a notable structural inertia will be the first to disappear. In this context, (Ning et al., 2010) argued that economic Darwinism would eliminate a large corporate board if a board with a large number of directors is harmful to firm value. Similarly, if large boards were better, economic Darwinism would eliminate small boards. To take into account this dynamic scenarios firm's corporate governance department requires method-based new decision support tools to determinate their optimal board structure. These tools must be able to predict the change in the board structure storage, and must allow exploring economic Darwinism.
3. Method and data
Multi-objective optimization is nowadays a familiar method for problems in topics such as product design, finance, facility planning, etc. In this work, we propose to apply a multi-objective genetic algorithm to the board structure trade- off problem. Traditional investigation in psychology and multi-criteria decision-making has underlined many features of the human decision-making practice which often involve the simultaneous concern of more than one objective function (Steuer, 1986; Miettinen, 1999). Every so often these objectives may be contradictory so that there is a trade-off between the benchmarks, and effective results will not be unique, but will remain among a set of ‘non-dominated solutions’ where any specific objective cannot be improved upon without worsening some other. Here we describe the fundamental trade- off of board structure in terms of the multiple objectives of firm performance maximization and agency cost minimization, and apply multi-objective genetic algorithm for obtaining the non-dominated solution set. The main reason for operating multi-objective genetic algorithms is their aptitude to find multiple non-dominated solutions in a single simulation. In view of the fact that the principal reason why a problem has a multi-objective design is because it is impossible to obtain a single solution which simultaneously optimizes all objectives, an algorithm that contributes a large number of alternative solutions sitting on or near the Pareto-optimal front is of good practical value. In this study, we used multiobjective optimization using multiobjective genetic algorithm function gamultiobj in MATLAB Global Optimization Toolbox with its default options. The data used in this study comprise firm performance measured by ROA and Tobin’s Q. As in Florackis & Ozkan (2009) we measure agency costs (AGENCY) via the asset turnover ratio stated as the annual ratio of total sales to total assets. The asset turnover ratio denotes “…. an inverse proxy for agency costs and [can be]
interpreted as an asset utilization ratio that shows how effectively management deploys the firm's assets … alow asset turnover ratio indicates poor investment decisions, insufficient ef- fort, and consumption of perquisites [and hence high agency costs]” (Florackis & Ozkan, 2009, p. 499). The four optimized board structures in this study are: (1) board size measured by number of directors; (2) board independence measured by the proportion of independent directors; (3) number of committee measured by number of board committee; and (4) committees independence measured by the proportion of independent directors in committee. Our sample includes 80 French firms belong to index SBF 120. The financial data are hand collected using annual reports, Paris Market Exchange and websites of selected firms. Our data covers 2001-2013 periods. Overall, we have 80 firms over a period of 13 years (1,400 observations).
2 It aims to arrive at conclusions about what things are good or bad, or what actions are right or wrong. In other words, a normative theory aims to discover what should be, and would include sentences like ‘companies should follow corporate governance standards’
or ‘managers ought to act in a manner to avoid conflicts of interests’.
5
The multi-objective board structure optimization problem:
Firms examine both the costs and benefits of large and more independent boards in determining optimal board structure.
Thus, optimal board is supposed to minimize agency costs. This can be achieved by reducing the opportunistic selection of financial accounting policies, and by increasing the credibility and accuracy of financial reporting. On the other hand, optimal board is also supposed to maximize shareholder wealth and firm performance. This leads to some unique trade- offs in corporate board structure. For instance, in order to minimize agency costs, boards must match resource dependency theory which suggests that companies are better off with large and more independent boards. Each new board member brings both expertise and access to resources. Having more board members, more independence and more monitoring committees would, therefore, provide the firm with greater knowledge and access to resources. These resources could comprise access to markets, access to new and better technologies... Large independent boards are more likely to have directors with greater range in culture and industry experience. This variety permits the board members to offer management with high quality advice (Zahra & Pearce, 1989). However, Lipton & Lorsch (1992) and Jensen (1993) claim that large and more independent boards and adding serve committees can be costly. Larger independent boards amplify operational complexity and increase the potential for disagreement among members, consequently leading to more director free-riding problems and internal conflicts among directors. This produces greater costs for the firm and affecting negatively firm performance. In this paper, we will try to determinate optimal board structure of board size, board independence, number of board’s committees and committees’ independence that simultaneously maximize firm performance and minimize agency cost. Consequently we set as targets (1) To ensure better firm performance, measured by ROA and Tobin’s Q and (2) minimize the of agency costs. Fitness functions (f1), (f2) and (f3) used in the optimization procedure was based on the multivariable nonlinear regression analysis model (MVNLR), (Tsoukalas & Fragiadakis, 2016) approximated by: f (βi xi, εi); where, εi is a random experimental error (e.g. a random measurement error) and ε2i = [yi – f (xi, βi)]. We used the Assistant in Minitab 17 which takes our candidate X variables and produces a regression model using stepwise regression which gives the minimum sum of squared errors (SSE), given by SSE= ∑𝑛𝑖=1W𝑖𝑖ε𝑖² Each component of the diagonal weight matrix W should, ideally, be equal to the reciprocal of the error variance of the measurement. Stepwise regression selects a model by automatically adding or removing individual predictors, a step at a time, based on their statistical significance. The end result of this process is a single regression model. We can control the details of the process, including the significance level and whether the process can only add terms, remove terms, or both.
4. Results and discussion
As a first step, we provide summary statistics for the numeric attributes of this research (Table 1). First, the ROA is around 4.67 % with a wide gap between the two extremities. It seems that in our sample firms are modestly performing.
Furthermore, we find that the average Tobin's Q in our sample is 1.36. This value seems to be comparable to that found by Boubaker, (2007) in the French context which is 1.99. The average agency cost is around 1,691 with a minimum of 0,033 and a maximum of 4,98. As regards to the corporate board structure, we found that, in average, board of directors is composed of 12 members with a maximum of 24 and a minimum of 4, which reveals the diversity and heterogeneity of the selected firms. However, this board of directors is considered with middling size compared to that in other countries.
In this concern, Chinese’s board size is on average around nine-member. The average of board independence is 0,53 % with a minimum of 0.2 % and a maximum of 100% which shows that the board of firms of our sample is dominated by independent members, which is consistent with the recommendations of the best practices. Finally, our sample has an average committee number of 3 and an average committee independence of 70%.
Table 1: summary statistics
Variable Obs Mean Std. Dev. Min Max
ROA 1040 0,0467 0,1351 -0,9068 0,999
Tobin’s Q 1040 1,369 1,413 0,0430 16,645
AGENCY 1040 1,6912 0,9605 0,0339 4,9887
BOARD SIZE (x1) 1040 11,8846 3,5859 4 24
BOARD INDEPENDENCE (x2) 1040 0,5313 0,1691 0,2 1
NUMBER OF COMMITTEE (x3) 1040 2,9327 0,7454 2 5
COMMITTEE INDEPENDENCE (x4) 1040 0,7005 0,1842 0,25 1
6 The Assistant uses standardized X variables because standardization removes most of the correlation between linear and higher-order terms, which reduces the chance of adding these terms unnecessarily. The final model is displayed in unstandardized (natural) units.
The final fitness functions are as follows:
f1: ROA = - 0,0145 - 0,01323* X1 + 0,0309 *X2 + 0,0912 *X3 - 0,0746 *X4 + 0,000325 X1^2 - 0,01173 X3^2 + 0,00428 X1*X4
f2: TQBINSQ = 2,045 - 0,04050 *X1 + 5,434 *X2 - 0,211 *X3 - 4,38 *X4 + 5,226 X4^2 - 7,216 X2*X4 + 0,298 X3*X4 Suppose g(x)= 1/(1+y(x)), maximize y(x) is equivalent to minimize g(x). .
f3: AGENCY = 1,667 + 0,0910 *X1 - 3,106 *X2 - 0,526 *X3 - 3,55 *X4 - 0,00629 X1^2 - 2,103 X4^2 + 0,0241 X1*X3 + 0,556 X2*X3
The optimization problem can be written as follows:
Minx F(x) = [g (x); y3(x)]
The boundary conditions for this optimization problem are as follow:
In France, Law provides that a board should be organized by at least three members and eighteen members maximum so we fixed 4<x1<18 and for the remaining variables we will take the lower and upper bounds of each; 20 % <x2<100%;
2<x3< 5 et 25 %<x4< 100% ( [4 0.2 2 0.25] et [18 1 5 1]). In order to apply the MOGA available in MATLAB Global Optimization Toolbox to our set of data, we authorized default options in its parameters: population size; number of generations; probability of crossover; probability of mutation. The outline of the optimization algorithm is given in Fig. 4.
Figure 1. Flowchart of optimization scheme based on combined MVNLR/MOGA algorithm.
The genetic algorithm handles a population of solutions in each of the iterations, instead of a single solution. Given that a population of solutions is treated in each of the iterations, the outcome of a GA is as well a population of solutions. When the optimization problem includes more than one objective function, the brief of locating one or more optimum solutions is called as multi-objective optimization. If the objective functions are contradictory, each objective relates to a different optimal solution. Therefore, there subsist a set of optimal solutions where again in one objective calls for a sacrifice in
7 some other objective. For a decision maker, identifying a number of optimal solutions becomes crucial and it also gives significant understanding into the decision thus offering several possible and valuable design solutions. Consequently, the best multi-objective optimization process is to catch the multiple trade-off optimal solutions with a wide range of values for the objectives and select one of the obtained solutions using higher-level information. If the objectives are conflicting with each other like in our case, it exist multiple Pareto optimal solutions. Therefore, a set of Pareto optimal solutions is expected when the multi-objective optimization problem involving firm performance maximization and agency costs minimization is solved. The gamultiobj solver attempts to create a set of Pareto optima for a multi-objective minimization.
These plots (fig5 and fig6) show the tradeoff between the components of f. It is plotted in objective function space. The Pareto plot displays two competing objectives. For this problem, the Pareto front is known to be disconnected. The solution from gamultiobj can capture the Pareto front even if it is disconnected.
Figure 2. Pareto-optimal fronts obtained by MOGA for ROA and AGENCY
Figure 3. Pareto-optimal fronts obtained by MOGA for Tobin’s Q and AGENCY
0.88 0.885 0.89 0.895 0.9 0.905 0.91 0.915 0.92 0.925 0.93 -6.5
-6 -5.5 -5 -4.5 -4 -3.5 -3 -2.5 -2 -1.5
Objective 1
Objective 2
Pareto front
0.2 0.25 0.3 0.35 0.4 0.45 0.5
-6.5 -6 -5.5 -5 -4.5 -4 -3.5 -3 -2.5 -2 -1.5
Objective 1
Objective 2
Pareto front
8 The results of the optimization appear in the following tables (Table 2 and Table3) containing both objective function values and the value of the variables.
Table 2. Results of the multi-objective genetic algorithm (using ROA)
Table 3. Results of the multi-objective genetic algorithm (using Tobin’s Q)
The results of this experiment display 18 rows. Each row represents one possible solution. A possible design of an optimal board structure can be gotten from the first raw of the Table 2 which seems to match a configuration of five members (more than 4) which confirm the efficiency of small boards and approve agency theory perspective (Yermack, 1996;
Gertner & Kaplan, 1996; Eisenberg et al., 1998; Denis & Sarin, 1999), all these five members have to be independent directors. This optimal board should also hold 4 committees and finally, 25% of committees’ independence is sufficient to
9 offer an optimal board structure that simultaneously maximize firm performance measured by ROA and minimize agency cost. To test the stability of the results, we carry out different experiment changing firm performance ratio (using Tobin’s Q) (Table 3). As we can see, the results are consistent throughout all the experiments, and the selected board structure seems to be the same. Thus, an optimal corporate board structure in France appears as one of these main possibilities, indicating that firms with fewer board members reap considerably greater rewards for their investors. After all, a board is nothing more than a group of human beings trying to work together to create the best results for the organization they are charged with directing and protecting. Human beings work best in groups of a certain size. In addition, the board independence is also relevant, showing that an independent board is a key determinant of board effectiveness. For a decision maker, when selecting board structure in practice, knowing a number of optimal solutions becomes important and it also gives considerable insight into the design thereby providing several feasible and useful design solutions.
5. Conclusion
In this work, we have presented the application of a novel method-based new decision support tools in selecting board structure in practice. The method permits one to have many criteria and can generate a Pareto-optimal front in a single simulation run. The problem addressed is that of the board structure trade-off in firm performance/agency costs. This problem is well known in the corporate governance literature, especially appearing in the agency and resource dependency perspectives as one requiring a possible multi-objective formulation. Furthermore we argued that board structure is pretended by dynamic scenarios such as the economic Darwinism. Thus, it will be necessary to deploy more and more decision-making tools. At the same time, it is clear that the tool developed here, will need to be used in more interactive ways by the decision-makers so that situations involving more variables can be sliced along appropriate sections for better visualization, thus empowering the decision-maker with new tools which is one of the modern requirements in corporate governance practice.
References:
Appay, B. (2005). La dictature du succès : le paradoxe de l'autonomie contrôlée et de la précarisation, Editions L'Harmattan, p. 262.
Beerel, A. (2009). Leadership and Change Management, SAGE Publications Ltd, p.288.
Booth, J. and Deli, D. (1996). Factors Affecting the Number of Outside Directorships Held By CEOs, Journal of Financial Economics 40, 81-104.
Boubaker, S. (2007). Ownership-control discrepancy and firm value: Evidence from France. Multinational Finance Journal, 11, 211- 252.
Braiotta, L. and Sommer, A.A. (1987). The Essential Guide to Effective Corporate Board Committees, Englewood Cliffs, NJ., Prentice-Hall.
Burt, R. S., Hogarth, R. M. &Michaud, C. (2000). The social capital of French and American managers. Organization Science, 11, 123– 147.
Chang, T.,Yang, S.,& Chang, K. (2009). Portfolio optimization problems in different risk measures using genetic algorithm. Expert Systems with Applications, 36 (7), 10529–10537.
Charreaux, G. (1987). La théorie de l'agence : une synthèse de la littérature de nouvelles théories pour gérer l'entreprise, Economica, p.151.
Chen, M. (2014). Using a hybrid evolution approach to forecast financial failures for Taiwan-listed companies. Quantitative Finance, 14(6), 1047–1058.
Dalton, D., C. Daily, J. Johnson, and Ellstrand, A. (1999). Number of Directors and Financial Performance: A Meta Analysis, Academy of Management Journal 42 (6), 674-686.
Dann, L., D. Guercio, and Partch, M. (2000). Governance and Boards of Directors in Closed-End Investment Companies, University of Oregon Working Paper.
Davalos, S.,Leng, F., Feroz, E.H.,& Cao, Z.(2014).Designing an if-then rules- based ensemble of heterogeneous bankruptcy classifiers: A genetic algorithm approach. Intelligent Systems in Accounting, Finance & Management, 21(3), 129–153.
Denis, D. and Sarin, A. (1999). Ownership and Board Structures in Publicly Traded Corporations, Journal of Financial Economics 52, 187-224.
Eisenberg, T., S. Sundren, and Wells, M.(1998). Larger Board Size and Decreasing Firm Value in Small Firms, Journal of Financial Economic 48, 35-54.
Florackis, C. and Ozkan, A. (2009). The impact of managerial entrenchment on agency costs: an empirical investigation using U.K.
panel data, European Financial Management, 15, 497- 528.
Gales, L. and Kesner, I. (1994). An Analysis of the Board of Director Size and Composition in Bankrupt Organizations, Journal of Business Research 30, 271-282.
Gertner, R. and Kaplan, S. (1996). The Value-Maximizing Board,” University of Chicago Working Paper.
Ghoshal, A.K.,Mukherjee,T.,& Dhar, S.(2011).An empirical study in Indian share market using neural network and genetic algorithm.
Asian Journal of Research in Banking and Finance, 1(1), 1–19.
10 Ha, S. (2013).Optimal insurance risk allocation with steepest ascent and genetic algorithms. The Journal of Risk Finance, 14(2), 129–
139.
Hadani, M.; Goranova, M.; Khan, R. (2011). Institutional investors, shareholder activism, and earnings management. Journal of business research, 64(12), 1352-1360.
Hannan, M.-T. et Freeman, J. (1989). Organizational ecology. Harvard University Press.
Heinrich, C.-E. et Betts, B. (2003). Adapt or die: transforming your supply chain into an adaptive business network, John Wiley and Sons, p.234.
Hoogs, B.,Kiehl, T.,Lacomb,C.,& Senturk, D.(2007).A genetic algorithm approach to detecting temporal patterns indicative of financial statement fraud. Intelligent Systems in Accounting, Finance and Management, 15(1–2), 41–56.
Huang, J.,Cai,Y.,&Xu,X. (2007). A hybrid genetic algorithm for feature selection wrapper based on mutual information. Pattern Recognition Letters, 28(13), 1825–1844.
Jensen, M., (1993). The Modern Industrial Revolution, Exit, and the Failure of Internal Control Systems, Journal of Finance 48, 831- 880.
Jensen, M.-C. (2001). Value Maximization, Stakeholder Theory, and the Corporate Objective Function, Journal of Applied Corporate Finance, 14(3), 8-21.
Kadushin, C. (1995). Friendship among the French financial elite. American Sociological Review, 60: 202–221.
Karimi, F., Dastgir, M., & Shariati, M.(2014).Index prediction in Tehran stock exchange using hybrid model of artificial neural networks and genetic algorithms. International Journal of Academic Research in Accounting, Finance and Management Sciences, 4(1), 352–357.
Leigh, W., Purvis, R.,& Ragusa, J.M.(2002).Forecasting the NYSE composite index with technical analysis, pattern recognizer, neural network, and genetic algorithm: a case study in romantic decision support. Decision Support Systems, 32(4), 361–377.
Lipton, M. and Lorsch, J. (1992). A Modest Proposal for Improved Corporate Governance, Business Lawyer 48 (1), 59-67.
Maclean, M., Harvey, C. & Press, J. (2006). Business elites and cor- porate governance in France and the UK. Basingstoke: Palgrave Macmillan.
Miettinen, K. and M¨akel¨a, M. (1999). Comparative evaluation of some interactive reference point-based methods for multi-objective optimisation, Journal of the Operational Research Society 50 (9), 949–959.
Morçöl, G. (2007). Handbook of decision making, CRC Press, p.638.
Nelson, C.-E. (2006). Managing quality in architecture: a handbook for creators of the built environment, Architectural Press, p.318.
Ning, Y., Davidson, W. & Wang, J., (2010). Does Optimal Corporate Board Size Exist? An Empirical Analysis. Journal of Applied Finance, 20(2), 57–69.
Oh, K.J., Kim,T.Y.,&Min, S.(2005).Using genetic algorithm to support portfolio optimization for index fund management. Expert Systems with Applications, 28(2), 371–379.
Pascale, R.-T. (2001), Surfing the edge of chaos : the laws of the jungle and the news laws of business, Leader to Leader, 20, 21-35.
Pfeffer, J., (1973). Size, Composition, and Function of Hospital Boards of Directors: The Organization and Its Environment, Administrative Science Quarterly 18, 349-364.
Rozsypal, A., & Kubat, M.(2003).Selecting representative examples and attributes by a genetic algorithm. Intelligent Data Analysis, 7(4), 291–304.
Sharma, B.,Thulasiram, R.K.,& Thulasiraman, P. (2015).Computing value-at-risk using genetic algorithm. The Journal of Risk Finance, 16(2), 170–189.
Shin, K.,& Lee, Y.(2002).A genetic algorithm application in bankruptcy prediction modeling. Expert Systems with Applications, 23(3), 321–328.
Steuer, R. E. (1986). Multiple Criteria Optimization: Theory, Computation, and Application (John Wiley & Sons, New York).
Tsoukalas, V.D. & Fragiadakis, N.G., (2016). Prediction of occupational risk in the shipbuilding industry using multivariable linear regression and genetic algorithm analysis. Safety Science, 83, 12–22.
Tufano, P. and Sevick, M. (1997). Board Structure and Fee-Setting in the U.S. Mutual Fund Industry, Journal of Financial Economics 46, 321-356.
Vafaie, H., & De Jong, K.(1992). Genetic algorithms as a tool for feature selection in machine learning, (pp 200–203).
Vasilakis, G.A.,Theofilatos, K.A.,Georgopoulos, E.F.,Karathanasopoulos, A.,& Likothanassis, S.D.(2013).A genetic programming approach for EUR/USD exchange rate forecasting and trading. Computational Economics, 42(4), 415–431.
Wu, C.,Tzeng,G.,Goo,Y.,&Fang,W.(2007). A real-valued genetic algorithm to optimize the parameters of support vector machine for predicting bankruptcy. Expert Systems with Applications, 32(2), 397–408.
Yang, J., & Honavar,V.(1998). Feature subset selection using a genetic algorithm. In H. Motoda, &H.Liu(Eds.), Feature Extraction, construction and subset selection: A data mining perspective (pp. 117–136).US: Springer.
Yermack, D. (1996). Higher Market Valuation of Companies with a Small Board of Directors, Journal of Financial Economics 40, 185-211.
Zahra, S. and Pearce, J. (1989). Boards of Directors and Corporate Financial Performance: A Review and Integrative Model, Journal of Management 15, 291-334.
