6.4.1 Introduction

The previous section explained that there are many differences among the cases. These differences warrant a different treatment, but how different should different cases be treated? The “one-metric approach” attempts to find an answer to this question.

The first step of this approach would be the following. Bank State aid decisions can contain a wide array of commitments and conditions with all different modalities. Would it be possible to use a single metric to capture all of these restructuring measures? If all the different restructuring measures can be captured into one single metric, then it becomes possible to compare the treatments in terms of that metric. What metric should be used? In other words: the treatments are to be compared in terms of what? One could choose a metric such as the severity (or harshness) of the restructuring measures. Since restructuring measures are often seen as “punishment”, it would make sense to look how severe the treatment is. The treatments can then be compared on the basis of their severity.1

The next step would be to do the same for the “cases”. State aid cases have many characteristics. The characteristics of the “case” are constituted by a combination of the characteristics of the bank and the characteristics of the State aid measure. In order to compare the cases, these characteristics of the cases should be combined into a single metric. One could choose for a metric such as the amount of ‘harm’ that the State aid has caused. If the amount of harm can be established, then the cases can be compared on the basis of this metric. In other words: it comes down to determining how different the cases are in terms of harm that they caused.

The final step would be to find out if the degree to which the cases differ from each other (in terms of the metric) can be related to the degree to which the treatments of those cases are different. The underlying idea is that the gravity of the case is related to the severity of the treatment. This follows from the principle of proportionality, which requires that the treatment is proportionate. In other words: banks where many things are wrong, deserve a harsher punishment (i.e. far-reaching restructuring measures).2

This line of reasoning can be illustrated by the following example. Assume that there are two cases (A and B) and that both the gravity of each case and the severity of the treatment can be quantified. The gravity of case A is 1 and the gravity of case B is 2. The severity of treatment A is 3 and the severity of treatment B is 6. See the following table.

There are two types of relations:

The principle of proportionality only concerns the relations [1:3] and [2:6]. The principle of equality concerns the way how these relations relate to each other, so [1:3]=[2:6] or [1:2]=[3:6]. The principle of equality works in two ways, which lead to the same outcome. It requires that the proportionality is the same for each case. So if treatment A is three times higher than the gravity of case A, then treatment B should also be three times higher than the gravity of case B. In numerical terms, [1:3]=[2:6]. Another way of putting it, is that the principle of equality requires that the degree to which cases A and B differ, should be reflected in the treatments. So if the gravity of case B is two times higher than the gravity of case A, then the severity of treatment B should also be two times higher than the severity of treatment A. In numerical terms, [1:2]= [3:6].

This example clearly illustrates that the one-metric approach can only work if the gravity of the case and the severity of the treatment can be quantified. Cases can only be compared in terms of a metric if a certain value can be assigned to that metric. Since this is not the case, the one-metric approach will not work in practice. In the following two subsections, this will be explained in more detail.