The Supervisory Capital Assistance Program-Part XIV

This is the fourteenth in a series of articles on the SCAP.

Category-wise review and assessment:  The SCAP supervisors applied a range of models and tools in carrying out the review and assessment of the submissions of the 19 BHCs.   We shall study, in detail, the category-wise review and assessment done in regard to the following:  first and second lien mortgages, credit cards and other consumer loans, commercial and industrial loans, commercial real estate loans, other loans, securities in available for sale and held to maturity portfolios, trading portfolio losses,  counterparty credit risk, pre-provision net revenue, and allowance for loan and lease losses.

First and Second Lien Mortgages:  Residential mortgages were an area where many of the BHCs took a big hit on their bottomlines.   The firms submitted information relating to this portfolio, the methods adopted to project losses, the important underlying assumptions in the methods adopted, etc.   Apart from this, at the instance of the supervisors, the firms also provided the following information about their residential mortgages:

1. Type of product
2. Loan to Value Ratio
3. FICO score
4. Geography
5. Level of documentation
6. Year of origination
7. Other relevant features

The agencies evaluated the first mortgages, home equity lines of credit, and closed end second mortgage products seperately.   They evaluated submissions of each firm in regard to the model, assumptions, and circumstances independently, as well as in relation to the peers in the group, to make any adjustments, if necessary.   The firms’ assumptions regarding prepayment of existing loans, and availability of new limits was normalized to make generally consistent across firms.   Each firm’s portfolios were analyzed by taking into account the unique features of the portfolio, and application of common loss estimation methodologies in the backdrop of industry-wide data.   Where any particular characterisitc of the loan portfolio like FICO, LTV bands, etc., provided any indication of defaults, they were applied more intensively to evaluate the submissions and make necessary adjustments to them.

Credit Cards and other Consumer Loans:  In case of credit cards, the methods used by the firms to project losses were in relation to the two macroeconomic scenarios.   Each of the firm’s results were benchmarked against historical trends in such portfolios comprising of loss, paydown/runoff, roll rates, utilization, etc.   Fhe firms submitted information called for about credit cards on the FICO scores, payment rates, utilization rates, geographic concentration, etc.

The supervisors, in addition to using existing models to evaluate the firms’ credit card loan loss projections and resources, also developed risk profiles suited to specific portfolios, that enabled them to make cross-firm comparisons.   That, in turn, revealed whether the firms had furnished reasonably accurate submissions.  It was found that, generally, the results obtained by the supervisors were relatively close to those of the BHCs.

Coming to the other loans, the bulk of which were auto loans, personal loans, and student loans, the firms submitted information on FICO scores, LTV, Term, vehicle age, and geographic concentratin.   The above data was examined in detail, alongwith the embedded components of each firm’s portfolios.   Further, historical loss experience, and performance measures were examined to arrive at the acceptable level of projected losses.

                                                                                                     To be concluded. 

Acknowledgement:  Adapted from the official document of the Board of Governors of the Federal Reserve System.