Credit Portfolio Reporting for Effective Risk Management [RMA Journal, The]

Financial institutions are becoming more effective at gathering and reporting portfolio-level data, but are they actively using the data to drive decision making and planning throughout the organization?

With all the regulatory buzz around the importance of portfolio reporting and stress testing, one might assume the increased focus on both has a direct link to more informed portfolio strategies and effective risk management policies. But that may not always be true.

Financial institutions are becoming more effective at gathering and reporting portfolio-level data, but are they actively using the data to drive decision making and planning throughout the organization, specifically in the area of credit risk management? To increase the effectiveness of portfolio management, financial institutions need to leverage the bounty of data produced by effective reporting and then translate that data into credit strategies and risk administration policies.

Let's step back and review the type of reporting, including stress testing, that every financial institution should be implementing. The need to understand current concentrations throughout the loan portfolio has driven most banks to invest in systems and processes for detailed segmentation reporting. Creating monthly reports that segment the commercial loan portfolio across loan type, collateral type, borrower risk rating, industry classification, geography, and other variables and comparing this information across time periods is the first step in developing actionable data.

To achieve this level of detailed reporting, each loan must be accurately coded and its information entered into appropriate systems for reporting purposes. Failing to gather the appropriate data at loan origination, and not updating the data as needed throughout the life of the loan, was a problem some banks discovered and corrected as regulatory pressure for segmentation reporting ratcheted up over the past few years.

Actionable Data as a Result of Segmentation Reporting

Effective segmentation reporting should be more than a way to look at the past; it should be used as a way to chart the future. It can serve as a road map for implementing credit risk management policies and practices. A good example of how to put segmentation data to work is to look at reports of credits by borrower risk rating (see Table 1 and the graphical representation of the segmentation report).

Closely reviewing and monitoring these reports over time can indicate whether your internal risk-rating system is 1) granular enough to provide meaningful data, 2) consistently applied throughout the organization, and 3) effective as an early warning system.

How can one segmentation report provide all of that insight? It can't. But by understanding the correlation between segmented risk-rating data and other data, you can learn a lot. It's the combination of data and subsequent data analysis that makes portfolio reports more than they appear at first glance.

First, is your risk-rating scale granular enough to provide meaningful data? Analyzing the percentage of borrowers in each risk-rating category will allow you to determine whether the rating scale and process you are using are due for an overhaul. An indicator that you may not be getting the best data from your risk-rating process is if 40% or more of your portfolio is lumped into a single risk category. Having a large concentration in a single category may indicate that the data being evaluated to determine a borrower's risk rating is not detailed enough to differentiate good credits from average credits.

While it's easy to identify the best- or worst-performing credits, it's often the credits in the middle that may need more attention. Having them all in a single rating category does not provide meaningful insight into your portfolio. Using this analysis to initiate a review and implement changes to improve your risk-rating process is an example of using portfolio reporting to drive changes in your overall risk management processes.

Second, is your risk-rating system being applied consistently throughout your organization? Segmenting your risk-rating reports further into geographic regions representing various markets, locations, or even branches within your organization can help you determine how your risk-rating system is being applied across the bank. If one region has the majority of its ratings grouped into two or three categories, while other regions have a distribution across categories that matches that of the portfolio, it may be that certain areas of the bank are not using the risk-rating system in the same way as the rest of the organization. This analysis may also help in determining if your risk-rating system uses the right mix of quantifiable data versus judgmental data.

Third, is your risk-rating system an effective early-warning system? This question leads us to portfolio stress testing in combination with segmentation reporting.

Portfolio Stress Testing

The next logical progression of portfolio reporting is stress testing. Today, banks are routinely performing the following stress tests:

* Collateral-value stress tests.

* Variable-rate impact analysis.

* Commercial real estate stress tests.

* Risk-migration reports.

Stress testing can be a useful management tool at both the individual loan level and the portfolio level. Individual segments of the portfolio should also be stress tested, especially if there are increased concentrations in the portfolio. Segmentation stress testing will allow for the identification of individual loans (particularly larger loans) that may need to be stressed at the transaction level to create awareness of potential risks. This report can be used as an early indicator for loans that may need to be monitored closely.

Collateral-value stress testing as an early indicator of potential problem credits is now commonplace at most banks. Most financial institutions capture data required for stress testing, such as the appraisal date and recent appraised value of the collateral. With several simple pieces of data, it is possible to recalculate loan-to-values in a portfolio-level report by doing an impact analysis of changes in collateral values. Having the most recent appraisal date, the appraised value, and the total debt on the property will allow for general stress testing.

Additional data can bring more depth and meaning to the collateral stress tests being performed. This data includes lien position, priority lien amount(s), allocations of collateral amount where cross-collateralization is present, appraisal methodology, and, in the case of commercial real estate and the income capitalization approach, vacancy rates, net operating income, and the historical discount or capitalization rate.

In Table 2, the balances in excess of the stressed collateral values can be evaluated as an implied unsecured or impaired balance subject to individual account review and analysis.

Performing a variable-rate impact analysis is also helpful in understanding the potential risk in a portfolio during times of interest rate shifts. While such shifts have not been a significant factor for financial institutions in the past two years, it is still very important for banks to understand the potential impact of future rate changes.

In light of the recent financial crisis and the impact to most lenders' CRE portfolios, regulators have increased pressure on financial institutions to perform account-level and portfolio-wide testing that stresses the interest rate, cap rate, and net operating income to determine the impacts to debt service coverage on CRE loans. If key elements of a typical cash flow analysis-effective gross income, total expenses, earnings before interest, taxes, depreciation and amortization (EBITDA), and total annualized debt service-are available, various scenarios can be produced and debt service coverage ratios can be recalculated at the portfolio level to identify loans at risk. If vacancy rates and expense items are captured, then stress factors can be even more specific to market conditions.

Add geographic identification and/or industry classification, and it's possible to target a portfolio segment more specifically. Add key elements of the debt service-outstanding balance, interest rate, term, amortization, rate-change date, floors and ceilings, and base rate index-and it becomes possible to recalculate debt service more accurately.

Another form of stress test that can help management identify portfolio shifts and potential at-risk credits is a risk-migration report that shows changes in balances and changes in risk ratings over multiple periods. These reports support calculation of normal migration and provide a way to determine the impact of a stressed migration as recommended by regulators.

Using Stress-Test Results In order to determine if your risk-rating system is an effective early-warning system, you must be able to create risk-rating migration reports that enable you to drill down to specific loans and identify borrowers on a potential downward trajectory. Monitoring those loans through risk-rating migration reports will provide insight into how accurate the changes in risk rating are in determining potential problem credits.

In addition, reviewing credits that are identified as watch-list loans and reviewing their individual risk-rating migration trends will help you determine which percentage of loans on the watch list would have been predicted based on periodic risk ratings.

In Table 3, you can see how combining segmentation results as part of various stress-testing scenarios can provide more granular, actionable information for a management team to use in future lending and policy decisions. All of the various forms of segmentation reporting and stress tests provide data that should be used when managing a portfolio. However, many bankers continue to view these reports as a regulatory requirement to be reviewed only periodically, rather than as an effective risk management and mitigation tool.

Bottom line, if you consider segmentation reporting and stress testing a regulatory burden rather than an effective way to inform ongoing monitoring and decision making throughout the organization, you are probably missing the point regulators are trying to make.

To achieve this level of detailed reporting, each loan must be accurately coded and its information entered into appropriate systems for reporting purposes.

While it's easy to identify the best- or worst-performing credits, it's often the credits in the middle that may need more attention.

Many bankers continue to view these reports as a regulatory requirement to be reviewed only periodically, rather than as an effective risk management and mitigation tool.

By Kimberly Songer*

*Warren Cole and B.J. Hall, senior risk consultants with Harland Financial Solutions, contributed to this article.

is director, Risk Solutions, Harland Financial Solutions. She can be reached at [email protected].