Skip to main content

Modeling bank loan losses in extreme events

This project is led by Dr. Sara Jonsson, who works at the Center for banking and finance at KTH.


Economic crises often unfold in a relatively short time period, effectively as extreme events. Banks need to be prepared to handle extreme events as a risk, especially since they have so much assets in lending. Extreme events can rapidly turn into loan losses for a bank, and as a worst case scenario create a ‘credit crunch’. Over the years, it has become evident that some banks are better able handle lending risks than others. For instance, Handelsbanken managed the financial crises of the 1990s and the 2010s very well. An explanation may be that banks have different models for credit risk evaluation, for example regarding credit evaluation procedures and the handling of payment defaults. Much points in the direction that a banks way of organizing its operations, decision making procedures, delegation of authority etc. has a large effect on the handling of situations where loans default. Research on credit risk evaluation and credit risk modeling is substantial. However, there is very limited research on intra-organizational factors that may influence banks’ ability to handle risks and loan defaults.

The purpose of this research project is to contribute towards an increased understanding of how the effect of extreme events on loan losses can be modeled.

Extreme events are unknown before they occur, otherwise they would not be extreme. For that reason, it is difficult to theoretically model the occurrence of extreme events. Instead, we focus on the operational modeling of the bank organization’s capability to cope with an extreme event. Our focus is on the loan losses suffered by a bank, since these are known to have caused state bail outs.