Implementation of a Credit Risk Rating Framework

Experience has shown than the behavior of SME corporates regarding their credit obligations depends heavily on their relationship with their bank, their peers, customers, region and other characteristics of the market in which they operate. As a result, an SME risk assessment model that works well for a particular financial institution might not work adequately for another, contrary to models developed for the retail sector or large corporates which can be fairly easily adapted to different markets.

Main Features

In addition, there are a number of further challenges which are posed for the SME sector:

  • Important information which is typically used for large corporates such as market price of issued credit derivatives, bonds, and equity is typically not available for SMEs.
  • Country accounting conventions need to be taken under account
  • Creative accounting can invalidate key model data
  • Lack of history on new/ modified companies
  • Limited default sample

At SYSTEMIC we have paid special attention to this sector which quite often comprises the largest portion of most banks assets portfolio, many of which are not entirely satisfied with the statistical performance of their existing SME rating models.

In order to address each of the above challenges as well as others, we offer:
  • Local analysis of actual obligor performance in the medium and long run. Use of best suited model from our own database, in the short run.
  • Localized accounting templates
  • Possibility to adjust financial information to address creative accounting
  • Creation of interim statements as well as future projections to address potential lack of history or modified companies
  • Use of macroeconomic risk drivers allowing to take under account the credit risk cycle and allowing to apply stress tests
  • Accuracy enhancement by the use of qualitative assessments for each company
  • Complete audit trail and application of “four-eye” principle allows to avoid potential mis-use of the model

The above results in a risk rating framework which is truly internal to each of our customers and delivers highly robust accuracy scores.