Highlights
At Issue
Recent corporate failures and scandals call for corporate governance reform. A lively debate developed on how to structure boards of directors in order to ensure effective decision-making. Which board structures serve shareholders' interests best? Do outsider-controlled boards outperform insider-controlled ones? Are two-tiered boards more effective than single-tiered ones?
Approach
To compare different board structures' effectiveness, the authors develop a theoretical framework identifying each board's possible outcome. They then conduct an experimental study which observes university students' behavior under different board designs. The experimental approach allows to identify which board structure features generate an efficient outcome, regardless of other factors that could influence decision-making.
Findings
The presence of outside directors on corporate boards results in better decision-making. In particular, the most efficient board structure is a single-tiered board with a majority of outside directors. The participation of outside directors improves the quality of corporate decision-making even when they are not the majority. The authors also show that two-tiered boards' investment decisions are too conservative. Experimental findings support a reform of corporate boards to include truly independent outside directors acting as "watchdogs".
Novelty
Typically, empirical studies relating board structure to firm performance rely on firm data. The main drawback of this approach is the difficulty of distinguishing the effects of the board structure from other institutional, economic and social factors that also affect firm performance. The paper’s experimental approach makes it possible to isolate the board structure's features that specifically enhance firm value.
Awareness of the importance of good corporate governance has risen after the collapse of Enron and other corporate failures in the United States and Europe. Shareholders, activists, government agencies, academics and practitioners all agree on the need to undertake corporate board reforms that will prevent malpractice while promoting economic efficiency. However, no agreement yet exists on which optimal board structure should be promoted.
Board structures vary considerably. A board of directors may include top executives - often referred to as “inside” directors - and outside directors who are neither employees nor stakeholders in the company. Two-tiered boards have one tier acting as management board and the other acting as supervisory board. Do these differences matter? Should we expect boards with a majority of outside directors, like those prevalent in the United States, to perform better than insider-controlled boards, typical of the United Kingdom and continental Europe? Are two-tiered boards, widespread in Germany and the Netherlands, more effective than single-tiered boards?
Empirical research on this issue has to reckon with the impossibility to distinguish the consequences of a specific board structure on firm performance from the consequences of other factors, such as ownership structure, industry factors, business cycle, etc. For instance, firms with two-tiered boards usually have a large, high-powered outside investor, and one cannot separate each feature's consequence on firm performance. The authors conduct a laboratory experiment so as to overcome this drawback. Experiments are artificial and not realistic, and are therefore not devoid of criticism; nonetheless, they are a powerful tool to investigate issues that real data cannot resolve, for the very reason that they create an environment isolating the effects caused by the factors of interest.
The authors analyze four different board structures: (1) a single-tiered board with a majority of inside directors; (2) a single-tiered board with an outsider majority; (3) a single-tiered board with just one insider; and (4) a two-tiered board. They develop a theoretical model of corporate board decision-making, and build an experiment with university students to test its predictions.
The model's board of directors has to decide whether or not to undertake a project. If the project is good the firm’s value increases, while it decreases if the project is bad. From the shareholders’ point of view, only good projects should be undertaken. Inside directors - managers - are conscious of a project's quality; it is in their personal interest, however, to accept all projects, regardless of their quality. On the contrary, outside directors - “watchdogs” - are not conscious of the quality of the project but would block any low-quality project.
In the first place consider the potential outcomes of a single-tiered board with a majority of inside directors. Insiders outnumber outsiders, and since it is in their interest to coordinate and vote to accept the project, all projects are undertaken regardless of quality. When coordination among insiders is not possible, insiders will provide honest recommendations and an efficient outcome is obtained, that is, only good projects are undertaken.
With a majority of outsiders, watchdogs can block insiders’ proposal. If insiders are able to coordinate and avoid revealing the project’s quality, outsiders will not be able to infer it. Since they want to avoid approving value-decreasing projects, they will reject all projects, both good and bad. If insiders cannot coordinate, they will reveal the project’s quality instead and an efficient outcome is obtained. A single-tiered board with only one inside director is a particular case of board with an outsider majority and the coordination of insiders. In this case, an efficient outcome cannot be achieved.
Finally, consider a two-tiered board, where one tier only includes inside directors and the other outside directors. The insider board votes first. If a majority of insiders reject the project, it is not undertaken. If a majority of insiders accept the project then, the outside board votes. The project is only accepted if a majority of watchdogs accepts it. With this board structure the predicted outcome is exactly the same as in single-tiered boards with a majority of outsiders.
These results show that a board's composition is quite important to ensure efficient decision-making. Are the results confirmed in an experimental setting? This is the object of the research's second step, where students are organized in groups of seven - a board - and each student is given the role of insider or outsider. Students acting as insiders and outsiders get paid small sums of money which reflect real life incentives of actual board members.
The board's members discuss and exchange information, and then the board takes a vote. The experiment's results show that the most efficient is the single-tiered outsider majority board with multiple insiders: 93% of the groups with this board structure achieve an efficient outcome - i.e., they accept a project when good and reject it when bad. Two-tiered boards are also quite efficient: an efficient outcome is achieved in 80% of the cases. However, these boards are also too conservative: they reject good investment projects too often. When there is a majority of insiders on the board, an efficient outcome is reached only in 62% of the cases.
Some useful lessons may be learned from this exercise. Boards need to include more than one inside director. A multiplicity of insiders is desirable to inhibit coordination and promote truthful revelation of insider information. The presence of truly independent outside directors is desirable even when they are a minority. Board structures should also vary according to the firm’s situation. High-growth firms may prefer single-tiered boards, while firms in mature industries where there are less attractive projects to be found may prefer a more conservative two-tiered board.
Report Source:"Board structures around the world: An experimental investigation", a Working Paper by: Ann B. Gillette (Coles College of Business, Kennesaw State University), Thomas H. Noe (Freeman School of Business, Tulane University), Michael J. Rebello (Freeman School of Business, Tulane University).
pp.51