3.5.2.3 External validity

It must be recognised that the prediction model developed in this study is based on a data set involving a selection of court cases from 2003 to 2013. I have not exposed the prediction model to a different external sample. In the current analysis, my intention was not to generalise the observed cases to all possible directors’ liability cases. Additional studies showing similar results are needed in order to be more confident that the obtained prediction rates prove to be stable. It is therefore problematic to generalise the findings of this study to other sets of cases in the past or in the future.

Moreover, I have not dealt with the question of how courts assess the extent to which ‘foreseeability of damage’ is apparent to a director. Case law only instructs an objective test, disregarding material conditions that may lead to a court’s finding that damage was foreseeable to a director. The model would gain more importance if there was a better understanding of the conditions under which the courts made such determinations.

Finally, Table 7 only recognises those factors that appear to have a bearing on the determination of directors’ liability in cases not involving subjective bad faith. I did not address the question of why these factors are important for reaching court decisions in such cases. The model would gain in significance if the ‘why’ could be answered to a sufficient extent. This however requires a normative validation.