The Use of Qualitative Indicators in Banking Rating Systems

The article compares internal rating systems of three banks from the German-speaking region, continuing with last year's research. In this paper a detailed analysis of qualitative indicators (soft - facts) is made. These qualitative indicators, as one of the two main components of banking rating systems have the wage of between 30% and 50% of the overall rating score. This makes this part of rating certainly important enough to be further researched.  The research is focused on the rating of business entities, more precisely the corporate, (especially limited liability companies or joint-stock companies). It does not deal with the rating of natural persons or non-profit organizations, municipalities etc. The procedure of collecting empirical data as well as data from relevant literature, their assessment according to the criteria of verifiability and relevance and the application of the induction method was used and a generalization of conclusions was subsequently made. The goal of this research was to find out if the structure of used qualitative factors (soft- facts) is similar or even the same across the rating systems included in the comparison and what weights of individual factors are used. The result of the research shows that two categories of qualitative indicators (soft - facts) are present in all considered rating systems: (i) quality of company’s management and / or strategy and (ii) market on which the bank client operates. (iii) Accounting or related indicators like information system or audit quality also play a significant role in rating systems. On the other hand, the use of the factor (iv) relationship with the bank (or similar) is quite different across the rating systems included in the research. The number and structure of guidance questions that help risk-management analysts determine indicator values also differ. In one case, there is an extensive catalog of questions with a standardized set of responses. In other cases, the number of questions is lower and each one has its specific variation of the predefined answers the analyst selects from.