Asset-by-asset approach

Traditionally, banks have taken an asset-by-asset approach to credit risk management.
While each bank’s method varies, in general this approach involves periodically
evaluating the credit quality of loans and other credit exposures, applying a credit
risk rating, and aggregating the results of this analysis to identify a portfolio’s
expected losses.
The foundation of the asset-by-asset approach is a sound loan review and internal credit risk rating system. A loan review and credit risk rating system enables
management to identify changes in individual credits, or portfolio trends, in a timely
manner. Based on the results of its problem loan identification, loan review, and
credit risk rating system, management can make necessary modifications to portfolio
strategies or increase the supervision of credits in a timely manner.
Banks must determine the appropriate level of the Allowance for Loan and Lease
Losses (ALLL) on a quarterly basis. On large problem credits, they assess ranges of
expected losses based on their evaluation of a number of factors, such as economic
conditions and collateral. On smaller problem credits and on ‘pass’ credits, banks
commonly assess the default probability from historical migration analysis. Combining
the results of the evaluation of individual large problem credits and historical
migration analysis, banks estimate expected losses for the portfolio and determine
provision requirements for the ALLL.
Migration analysis techniques vary widely between banks, but generally track the
loss experience on a fixed or rolling population of loans over a period of years. The
purpose of the migration analysis is to determine, based on a bank’s experience over
a historical analysis period, the likelihood that credits of a certain risk rating will
transition to another risk rating.

Default probabilities do not, however, indicate loss severity; i.e. how much the
bank will lose if a credit defaults. A credit may default, yet expose a bank to a
minimal loss risk if the loan is well secured. On the other hand, a default might
result in a complete loss. Therefore, banks currently use historical migration matrices
with information on recovery rates in default situations to assess the expected loss
potential in their portfolios.