3.4.1 Which behavioural and contextual legal case factors have a significant effect on directors’ liability?

Table 1 shows the bivariate associations between the legal case factors under study and the outcome of a case. Pearson’s chi-square test was used to investigate whether the relationships were statistically significant (sig. ≤ 0.05).56 Cramér’s statistics (V) was applied to obtain an idea of the strength of a potential association.1

The research results show that there is a significant relationship between the factor of ‘unreasonable risk taking’ and the case outcome of a director being or not being found liable. The strength of the relationship is however weak (V = 0.214). The same can be said for the ‘policy failure’ factor (V = 0.164). These findings find support in case law, where directors are given discretion with regard to business judgement. The finding even suggest that courts ostensibly protect risk taking that prove (in hind sight) to have been excessive risk taking or business strategies that prove (in hind sight) to be failures. In other words, directors are actually unlikely to be deemed liable on the basis of these factors.

There is also a significant relationship between ‘fraud’ and case outcome, but the strength of this association is also weak (V = 0.255). A plausible explanation is that fraud, in the majority of observed cases, was not or not solely caused by the directors under review.

Of the legal case factors that were significantly correlated with the case outcome, two have a strong association (V ≥ 0.5): ‘foreseeability of damage’ (V = 0.545) and ‘subjective bad faith’ (V = 0.556). I also found a medium strong association between ‘norm violation’ and case outcome (0.3 ≤ V ≤ 0.5). It is therefore not surprising that closer inspection of these three legal case factors reveals interesting observations.

Table 2 shows that the director was found liable in 89% of the cases involving a ‘norm violation’. In other words, in 11% of the cases involving a ‘norm violation’, the director under review managed to successfully advance circumstances in his defence to refute the claim.

Furthermore, 96% of the cases involving ‘foreseeability of damage’ on the part of the director, resulted in the director being held liable. This research result indicates that, whenever a court assumed that the damage to creditors was reasonably foreseeable, the chance of a director to successfully advance circumstances to refute the claim was very unlikely.

Strikingly, Table 2 reveals that directors were adjudged liable in all cases involving actions by the director in ‘subjective bad faith’. The legal case factor of ‘subjective bad faith’ explains the case outcome (in one category) perfectly (100%). In other words, when a court determines that a director’s action involves ‘subjective bad faith’, other factors seem irrelevant: the subjective bad faith action is alone sufficient to rule that a director is personally liable. Moreover, paragraph 3.3.4 raised the issue of whether there being problematic inter-correlations between the predictor variables used in this study. If, for example, inter-correlation exist between the factors of ‘subjective bad faith’ and ‘foreseeability of damage’, one of the factors might be redundant.2 There were however no concerns with regard to multicollinearity. The factors ‘subjective bad faith’ and ‘foreseeability of damage’ seem to be relatively distinct, independent variables.3 Because the sample of cases under study involved quite a large number of ‘subjective bad faith’ cases (one third of the sample involved ‘subjective bad faith’ actions on the part of the director), I wanted to understand this factor better. Consequently, I also examined the factor ‘subjective bad faith’ as a dependent variable.