Unpacking board model choice in Europe: a multi-methods approach

• Department of Law

Join us for the November session of the Law Department faculty seminar series!

What explains the internal governance structure of organisations? More specifically, why do companies design their corporate governance in a particular manner? This paper uses firm-level data from fourteen European countries that give public companies a choice between multiple board models, notably a one-tier model with just one board of directors and a two-tier model with both a management and a supervisory board. It analyses the data with a multiple-method research strategy: a conventional regression analysis, a 'random forest' machine learning method and a qualitative interpretation of the findings. The random forest analysis is a key contribution of this paper, as it has not been applied in this field of research and as it can be particularly helpful in understanding a large number of possible reasons, forms of non-linearity and other complexities. The paper distinguishes between micro, meso, and macro reasons of corporate governance choice. It finds that all of these reasons play some role in explaining the board choices of individual companies: thus, it matters what is generally happening in a particular country (i.e. the macro-level), what divisions may exist between different groups of companies (i.e. the meso-level) and what is endogenous at the firm level (i.e. the micro-level).