How Risk Management Is Changing

September 24th, 2012

One inescapable truth about the post financial crisis: far more sophisticated, future-focused risk management is required going forward. New capital—and good business sense—demand it.

human-camera.jpgWhat does this have to do with credit culture, often associated with the practice of making sound loans? A great deal, at least as banking evolves beyond the financial crisis. To accomplish this requires a changed mindset, and credit culture is largely a mindset. Despite its traditionally esoteric nature, credit culture must not only be felt but enumerated. No longer should the former Supreme Court Justice Potter Stewart’s famous line about knowing obscenity when you see it be operative in defining your bank’s credit culture. To define credit culture, this leads us to the spectrum of qualitative versus quantitative, concepts known to every lender and credit officer. We believe it’s time to take this dichotomy beyond classic credit analysis, which most of us think of when hearing those terms, and project it onto the broader spectrum of optimal, future bank-wide portfolio risk management.   

Qualitative analysis implies subjective judgment—and in the greater scheme of things can also encompass the transactional, tedious loan-by-loan assessment of risk. We know transactional risk sensitivity is particularly important in relationship to the size of the bank. If one big loan goes down, that can have a huge impact on a small bank, for example. On the other end of the spectrum, which we can label quantitative, we have aggregated portfolio risk assessment, often combined with financial models used to predict outcomes. Many believe excessive dependence on modeling lead to the financial crisis in the first place.

Therefore, one could argue that on one end of the spectrum of optimal credit risk management, we have transactional risk (individual loan analyses and servicing) and on the other end, aggregated risk (modeling and forecasting). Worshiping exclusively on either end of the spectrum is problematic. Understanding only individual loan risks is like losing sight of the forest for the trees. And at the other end, we all know models are only as good as the integrity of what informs them (i.e., the transaction-based discreet data points, tediously mined from underwriting). There’s a sweet spot somewhere between the two polar opposites. But in the end, there must be a bottom line, a sum of all the parts to accurately project a bank’s risk profile, both present and future. Each stakeholder needs that profile depicted in summary fashion.              

How does one accomplish this blending of the theoretical qualitative and quantitative disciplines? Ideally, a bank should begin with an underwriting process that ensures adherence to its policies and guidelines—while cataloguing exceptions. Concurrently, it needs to have a complementary process that aggregates the individual loan data, and performs multiple portfolio-wide tasks, including stress testing, and calculating the allowance for loan and lease losses (ALLL), risk grade migration and potential credit losses (beyond the reserves), etc. Products have been developed to perform these tasks, on both ends of the spectrum, and frankly at costs far less than what some of the national providers of analytic tools are quoting.   

Lastly, much of credit culture jargon is historically, or backward- focused. That leads to statements such as “We grew this much in loans,” or “We had this percentage of past dues,” or “Our yield improved to this.” Regulators and investors read the call report. They are focused more on the future, as bank boards and management should be.  Boards need to ask questions such as, “Where are we trending in the various quality, growth, and profitability measures?” or “Where’s the next bubble in the portfolio?” or “Where will our non-performing loans likely be in one year?”

Success in banking going forward will be defined largely by institutions’ shifting from this traditionally historic to a forward-looking focus—and not just within a strategic planning or budgeting context—but embedded in the credit culture itself.  Simply put bank management and boards will have to be as conversant in PD (probabilities of default) and EL (expected losses) as traditional acronyms such as DSC (debt service coverage) and LTV (loan-to-value) ratios. With the widely anticipated high level of bank mergers and acquisition activity in the next few years, an additional premium will be placed on being informed and credible with these more macro and predictive concepts. They don’t replace the old reliable transactional measures; they just partner with them to quantify the risk going forward. The investors, the regulators—and common sense bank management—will mandate this.  


David Ruffin is a director at Dixon Hughes Goodman, LLP (DHG). Prior to DHG, Mr. Ruffin was co-founder and chief strategy officer of Credit Risk Management Analytics, L.L.C. Credit Risk Management was founded in 1989, and in 2015, it merged with Upland Analytics.