Management Models and Evaluation of Reputation Risks

Abstract

The article justifies the necessity to develop an analytical basis, management models and assessment of reputational risks. The complexity of the analysis of this category of risks associated with the presence of parameters that differentiate reputational risk from a number of other risks.

The effective management of reputational risk requires constant increase of transparency in reporting, which helps to strengthen the trust of stakeholders by providing reliable, timely and representative information about business. The reputational risks can be identified by two main corporate reporting functions: first, it corrects the expectations of stakeholders, showing how accurate the previous estimates were provided, for the second allows managing further information expectations.

Modeling the level of reputational risk and the magnitude of losses after risk event is an effective tool for taking management decisions by risk management units. Existing methods for the analysis do not take into account the whole range of factors and do not allow to comprehensively assess the consequences of reducing / losing business reputation of banking institutions.

The paper proposes a comprehensive methodology for assessment the level of reputational risks and the size of losses of banking institutions, that allows to combine an expert assessments and a statistical information about incurred losses (loss values), and also proposes the modeling of cause-effect relationships. This model based on the Bayesian belief network and Theory of Fuzzy Sets.

The advantage of such approach is on the possibility to evaluate the probability of some risk events based on the Bayesian theorem, that is, only on expert knowledge, and others based on empirical data on losses, if their volume sufficient for modeling purposes.