Falling asset values, deteriorating credit quality, and a reduced tolerance for risk have dimmed new loan appetite for many community banks and credit unions. Effectively assessing borrower and portfolio risk is more important now than ever before, driving everything from relationship and loan pricing to determining the reserves needed to mitigate economic fluctuations.
Many credit risk rating models today are manual and subjective in nature. Not only are they ineffective, but they’re no longer needed given the lending and risk solutions on the market today. These solutions provide a faster, more efficient way to evaluate risk, saving both time and money for the financial institution.
Let’s look at some of the strategies and objective data these types of solutions can provide to better and more efficiently assess and mitigate credit and portfolio risk.
The primary weakness of manual risk rating models is the granularity of the data. Many final ratings are judgmentally and manually completed on a spreadsheet that limits the user to reviewing the topics for consideration and picking what he or she thinks provides the best fit. This creates an issue with consistency in auditing results. Credit managers may weigh the statements differently, resulting in varied outcomes and reducing senior management’s confidence in the system they use to determine appropriate reserves and pricing.
Consideration should be given to breaking down the model to more specific questions for more consistent audit and justification of risk factors. The perception may be that more questions mean longer processing, but this is not the case. Ratings process faster, saving time spent determining the rating value and reducing the need for justification commentary.
Some judgmental elements of a rating system cannot be avoided, but the goal should be to minimize them. Risk models and solutions should be primarily objective, utilizing data, so that any two individual raters will make the same assessment on each factor. Vague metric and ratio descriptions should be replaced with specific data elements derived from the financial institution’s lending and risk software solutions. There will still be some subjective factors used in assessing the overall risk of a credit and relationship, but the goal should be to use more known and verifiable values.
Manual risk rating processes limit the financial institution to gathering information which it deems “predictive” of overall risk without accounting for supplemental objective data. Risk factors are strongly correlated, which means that risk models in today’s environment should couple data from lending and risk solutions to effectively verify the overall risk prediction. Technology is now able to automate many of these steps, saving valuable time and providing consistency to the risk stratification of the portfolio.
With the lending and risk solutions available today, quality risk analysis is achievable and made more efficient than ever before. Rather than looking at the initial rating for pricing purposes, or periodically as part of overall portfolio monitoring, having objective and consistent risk models in place allows for greater consistency in overall lending strategies. This provides deeper views into the portfolio for industry comparisons across peer groups to view and predict specific borrower and overall portfolio performance. Allowing for views of subsets of the data improves the overall quality and effectiveness of the analysis results.
A key to the effective risk rating of credits and the overall portfolio is consistent validation of the models being used. This is achieved by periodically taking a subset of borrowers and measuring them against the current risk model to understand the effects improvements to the risk rating models have on the portfolio in general, or on a select concentration of the portfolio. This allows the financial institution to stay ahead of the curve and take action to mitigate risk overall. This simple step will project the aggregate and concentrated portfolios’ potential migration period over period, and it will also help to predict risk migration in dollars to project possible changes in reserve needs.
Many small to medium-size financial institutions have traditionally utilized risk models that have been created in-house, as it’s been somewhat simple to create a combination of subjectivity and objectivity in the model. Technology powers objective versus subjective data and puts the mountains of borrower data available to work. The key is to identify the data that is most effective in predicting risk in your market and financial institution. If needed, seek the assistance of professional risk model builders and leverage their expertise to fine-tune your current model to ensure that it is sustainable going forward.
Data quality is your most valuable asset for effective credit and portfolio risk analysis. Lending and risk technology offered today are key to gathering data in a consistent and efficient manner, so that you are assessing risk from the most objective view possible – one that minimizes the judgmental elements of the manual risk rating system.
You’ll get the true story of borrower performance, making it easier and safer to provide access to capital for a wider range of qualified borrowers, thereby increasing your loan volumes in today’s challenging environment.
Who We Serve
What We Offer