Clustering Austrian Banks’ Business Models and Peer Groups in the European Banking Sector

As the European banking sector is becoming increasingly intertwined, the degree of interdependence is also rising. Consequently, it is key to conduct comparisons for a timely identification of emerging patterns of this development. Furthermore, the product range of banks has expanded so that heterogeneity across the banking sector has also been growing rapidly. This rising heterogeneity makes it increasingly impractical to carry out comparisons on an aggregate level. A more efficient approach is identifying one or ore ”common denominators” of similar banks and establishing groups of banks which share this (these) common denominator(s). In this paper, we consider the business models of banks as one such common denominator, which can be described by a set of variables. These ariables span a high-dimensional space where each bank represents a point, which can be measured by a statistical distance. Points close to each other may constitute a group, while points distant from these points will not belong to that group. Therefore, the objective of this study is, on the one hand, to define an efficient set of variables correctly reflecting the business models of banks and, on the other hand, to find subsets of high similarity. By applying statistical clustering techniques we aim to understand banks’ business models, thereby gaining new insights into the design of the European banking sector and, in particular, identifying peer groups relevant to the top Austrian banks. Assessing the distribution of risk and identifying certain business patterns within those groups allows a meaningful ranking of Austrian banks in comparison to their European competitors.2 The analysis in this paper is conducted on the basis of a purely quantitative methodology and the results should be interpreted accordingly.