Government Compliance
BASEL II - Probability of Default Analysis - Analysis to help you become BASEL II CompliantOCC Documentation and Validation Analysis - Analysis to help you become OCC Compliant
Sarbanes-Oxley Compliance - Implement statistical-based credit scoring for objective decisioning
BASEL II - Probability of Default Analysis
Becoming Basel II compliant is not an easy task and can be a huge undertaking, if a financial institution does not have the expertise or resources to meet the requirements. Meeting the demands of Basel II is a continuous challenge. But with PredictiveMetrics (PMI) performing the analytics, initial and ongoing compliance becomes easier.Many financial institutions are employing PMI to analyze, measure, and turn data into knowledge-based solutions, to improve financial performance and to help them meet the Internal Based Rating requirements of the Basel II Capital Accord.
Basel II has many businesses using statistical-based techniques for evaluating their credit and risk management practices. These forward thinking companies are utilizing PMI’s analytical solutions to integrate Basel II standards into their risk management strategies.
Becoming Basel II Compliant with PredictiveMetrics
PMI helps financial institutions become Internal Rating Basel II compliant by conducting a Probability of Default Analysis to help them meet their regulatory capital requirements. The goal is to develop a consistent methodology for estimating Basel compliant probabilities of default. The Basel requisite determines a financial institution’s capital requirements and overall corporate rating, which directly impacts their cost of funds.With PMI performing the analysis:
1. Capital allocation is more risk sensitive;
2. Operational risk is separated from credit risk, and both are quantified;
3. Economic and regulatory capital are more closely aligned, thereby reducing the scope of regulatory arbitrage.
When performing a model validation it is important to separate the new from existing customers due to the differences in payment behavior and the additional internal data that is available and can be analyzed on existing accounts.
Steps to Model Validation
When PMI validates a financial institution’s models, the following steps are performed:1. PMI prepares data requests covering a company’s internal data as well as data from each of your company’s chosen credit bureaus, gathering all of the data necessary to conduct the validation and provide the requisite analytics.
2. PMI match merges and arranges this data, ensures the quality of the data, and then enters the data into a custom designed analytical database of form and format that is most suitable to conduct analytics on the models.
3. PMI ranks the accounts from highest risk score to lowest risk score for each model to analyze the predictiveness of the bad credit definition chosen by the company.
4. PMI produces capture rate and bad rate tables and graphs across the distribution of scored accounts with the following corresponding performance statistics: Predictiveness Index (PI) and Kolmogorov-Smirnov (KS) statistic.
5. PMI presents the validation results to management and provides its recommendations regarding the continued use, possible improvements and applicability of the reviewed models.
PMI also reviews model documentation to make certain that the procedures for model development are being properly documented, and in the event that a key modeler were to leave that their work can easily be continued by a well qualified individual.
Statistical models tend to deteriorate over time. The most cost-efficient method of safeguarding against deterioration in model performance is to have the models’ coefficients re-fitted or re-estimated on a regular basis, thereby maintaining their predictiveness. PMI ensures model stability through ongoing revalidation.
Leading auditors have given statistical-based Credit Scoring the “stamp of approval” for helping companies implement non-subjective, consistent rules and policies. Public companies are turning to PMI for the development of statistical-based credit scoring models and replacing internally developed or acquired software that provides subjective judgmental/rules-based risk assessment. Statistical models are objective decision tools that aid in the SOX compliance mandate.
The CEO and CFO are no longer the only ones accountable for SOX compliance. Criminal prosecution for violating SOX is cutting across all revenue recognition functions, from sales to finance to credit. The credit manager often has responsibility over managing, accounting for and reporting significant assets and liabilities of the company including managing an accounts receivable portfolio often valued in the millions or hundreds of millions of dollars. Many companies have instituted a policy of upstream compliance requiring that credit personnel certify to the accuracy of the assets under their control. In particular, as the aging and collectability of the accounts receivable are many times measured by DSO and the bad debt reserve therefore having a significant impact on a company’s financial reporting to the public markets.
These calculations impact revenue recognition, and if not properly done can trigger an SEC investigation which, in a worst case situation, might require earnings to be restated. Given this, the credit professional must be concerned with SOX compliance.
BACK TO TOP
