April 24, 2012

Credit Union Management magazine’s Web-only “Loan Zone” column runs the fourth Tuesday of the month.

Many credit unions understand the need to identify and quantify risks within their loan portfolios, but such scarce resources as time, staffing or expertise can derail the process before it even begins. Executed properly, a risk analysis on your loan portfolio can have significant benefits and, with the right guidance, this seemingly insurmountable task can be achieved. Although hiring help is not always the answer, in our case, it was exactly what we needed.

In early 2009, our board and management grew concerned with the level of delinquency and number of charge-offs that were surfacing month after month. Additionally, there were signs that the economy could continue to shrink, which would further exacerbate these issues. No longer wishing to be surprised each month, our credit union’s leaders wanted to know how much of our portfolio was at risk of default, and what the potential losses could be on those loans.

These were valid questions but, at the time, the answers were difficult to obtain. What we did know was that attempting to analyze risk across our entire portfolio would be a challenge, given the thousands of loans on our books and our limited experience with risk analysis. We slowly realized that we didn't have the resources to take this project on manually.

The next few months went by rather quickly as we began our search for some answers. We reached out to some neighboring credit unions and colleagues to see what they were doing to manage their risk. Fortunately for us, several of our sources told us the same thing:

“Talk to a company called Twenty Twenty Analytics,” they said. “They’ve helped us do the things you’re talking about and we were really happy with the results.” So after numerous recommendations, it seemed like we had found our solution, and because we needed to move forward, we made the call.

Not long after, we met with a director from Twenty Twenty. He was very informative and talked about a variety of ways we could look at the risk within our loan portfolio. He pointed out that risk comes in many forms and each portfolio is unique, so there is no single formula or set of factors that can be universally applied to analyze risk. In other words, analyzing a loan portfolio is a custom job.

He went on to discuss how Twenty Twenty risk rates each loan, as well as assesses collateral values. This is how they go about quantifying risk and, because their process is automated, it drastically cuts down on the time it takes to accomplish this type of review.

Needless to say, we were sold, and within a few days of that first meeting, we were sending our raw system data to them for compilation and analysis. This transfer occurred from our core system to their secure website, and it was a fairly quick and straightforward process. In a relatively short amount of time, we received back volumes of information in the form of a data file and a report that we could easily navigate and use.

The final report graphically depicted and detailed the various risks within our portfolio, including analysis on loans at risk of loss, concentration risk, default risk, and the allowance for loan losses, to name a few. The data file served as the backup for the report, and it allowed us to dissect and drill down into different groups of loans and members for further review. It was at this point that we began finding our answers and, perhaps more importantly, managing our risk.

As we worked through the reporting and details of the loan data, we started to find that the true value of the information went far beyond our initial expectations. We were now looking at our whole portfolio and could see that, with the bad, there was also some good. The reports allowed us to separate out and initiate actions on specific groups of risk-rated loans. For example, we were able to target segments of our higher-risk loans for risk mitigation activities like reducing credit limits. Although not the most popular endeavor, this was a vital step in lowering the credit union’s overall risk level.

In the months following, we frequently talked with Twenty Twenty, and each time the company helped us dig deeper into the data. We began to better understand what attributes made up a good loan, how our strategies were impacting our portfolio’s performance, and where our new opportunities and emerging problems existed.

Most importantly, we were finally able to quantify risk in terms of dollars, which was something we could not have done previously. For instance, the report aggregated the uncollateralized portion on all of our high-risk loans, and this was significant because this number represented the risk within our portfolio that we needed to be concerned with – especially as it related to our net worth.  Additionally, we were able to leverage and use this number in other internal models and reports, so we really made the most out of the data.

Overall, the loan portfolio risk analysis from Twenty Twenty empowered us to not only find the answers we were looking for, but to identify new insights and opportunities that might have otherwise gone unnoticed.

If the last few years have taught us anything, it’s that we need better systems in place to understand, evaluate and act on risk. An effective risk analysis on your loan portfolio should not only help you understand your risks, but allow you to proactively make decisions and set strategies that improve your credit union’s financial health and performance.

Brad Stewart is enterprise risk manager for $843 million IBM Southeast Employees’ Federal Credit Union, Boca Raton, Fla.