The Power of Data as Rising Interest Rates Loom

By Justin B. Bakst06.29.2015

The NCUA has clearly outlined the supervisory priorities for 2015 and interest rate risk remains at the top of the list.  The message is clear; the buildup of potentially volatile member on-demand balances, as well as increased fixed rate loan activity over the last several years has created a heightened level of uncertainty for many credit unions.

The questions remain:

  • How will depositors react as the FOMC contemplates its first rate hike this decade?
  • Will mortgage activity grind to a halt, significantly impacting loan prepayment activity?
  • Will dividend payments need to move in lockstep with short-term rates to remain competitive and hold current share balances?

The answers to these questions will significantly impact credit union margins and liquidity as rates rise.

But there is not a “one-size-fits-all” answer to the potential rising rate exposure in the industry.  Each credit union is unique, with a distinct balance sheet composition and varying degrees of interest rate and liquidity risk.

Much of this risk is driven by assumptions regarding member behavior as rates move higher.  We see this projected member behavior vary across different markets and segments throughout the country, creating a wide spectrum of exposure levels.

Capturing these interest rate risk complexities begins with “best-in-class” modeling and measurement, while well-developed policies and a continuous monitoring process remain paramount to understanding risk levels and tolerances. A core component of this process is the use of existing data to better understand member behavior patterns. Data analytics, a convergence of data trends and customer/market analysis, provides meaningful opportunities for credit unions of all sizes to more effectively drive earnings, as well as manage the bubbling interest rate and liquidity risks inherent in all credit union balance sheets.

How Well Do You Know Your Members?

The amount of member data housed in systems throughout your organization is massive. According to a recent study, business data across all industries doubles every 1.2 years. This volume and velocity of data provide insights into trends previously difficult or cost prohibitive for credit unions to analyze. As the cost of storage and tools has plummeted, credit unions have begun leveraging this data to better understand member behavior patterns. This data includes customer banking trends, nature and size of relationships and demographic analysis—all of which impact risk profiles.

 

 

Member Deposit Behavior Patterns At a Glance

Betty-Ann: 55-year-old Betty-Ann has large multiple relationships with the credit union She has a share draft account that she opened nine years ago with an average balance of $12,000. She has $150,000 in a three-year certificate and a premium money market deposit of $18,000. She financed her house through the credit union and has seven years remaining on her mortgage. Betty-Ann is a “core member” of the credit union.

Frank: 37-year-old Frank has one account with this same credit union, a $90,000 premium money market. He opened the account in December 2009, in the midst of the financial crisis. Although the account is currently paying a dividend of 0.40%, it initially was paying 2.00%. Frank is a flight risk.

When interest rates move or market dynamics change, Betty-Ann and Frank will likely react very differently. In fact, the entire spectrum of your unique member base will have distinct and divergent behavior patterns. Unfortunately, few credit unions are performing analytics that capture this dynamic and even fewer are incorporating these behavior patterns into their interest rate risk and liquidity risk models.

Deposit Risk Factors—Where to Start

With seemingly endless amounts of member data, where do we start to better understand rising rate risk factors and separate the “Betty-Anns” from the “Franks”? A practical and pragmatic way to approach this challenge is a quantitatively robust historical lookback at each account, which many refer to as a core deposit study. Many credit unions are examining the statistical relationship between deposit accounts and dividends paid, balance inflows and qualitative measures including member “stickiness” factors and local and regional variables. This type of analysis can begin to separate the core members and the flight risks. Credit unions are utilizing this data to help form key assumptions in the interest rate risk and contingency liquidity model. For many credit unions, these assumptions can be the difference between an asset sensitive and liability sensitive interest rate risk profile and/or a well-executed growth plan versus a funding shortfall.

Beyond the “Math”

Although a core deposit study is a start, risk managers also need to think beyond the quantitative analysis. Too many credit unions perform this sort of analysis solely for the NCUA and move on. This approach is insufficient; a best practice deposit study helps inform management of key risk factors and identifies challenges and opportunities inherent in the deposit base, including new product opportunities and future pricing recommendations.

Isolating Loan Prepayment Risk

In addition to deposit uncertainty, most credit unions question how mortgage loan activity will react as interest rates rise. Many believe prepayments will dramatically slow, resulting in longer average lives. The utilization of existing data and data analytics can provide further insight. How many credit unions actually measure prepayment activity today? Unfortunately, most credit unions do not quantify historical prepayment speeds and differentiate between additional principal payments (curtailment) and refinance activity. Many are also unable to quantify the original loan size, date the loan was originated (vintage) and specific customer demographics.  These data points will clearly help to determine exposure levels. Customers with different vintages, payment history and demographics will all prepay mortgage loans distinctly with or without a corresponding change in interest rates. I find it curious that many credit unions are troubled with “tomorrow’s” rising rate risk factors, but most are not measuring these same factors today.  Data analytics can assist in this process.

Start Mining Today

What do you think of when you hear the word mining? Many of us imagine the great gold rush or the extraction of other valuable minerals from the earth including coal or oil. This type of mining can be time consuming and potentially costly but the payoffs are tremendous. The payoffs in data mining are no different—the goal is to extract valuable information to help us run our business more effectively. Credit unions can utilize existing data to enhance the risk management process and address the NCUA’s hot button issues, including rising rate exposure. In addition, data analytics can assist in the strategic planning process as well as marketing and sales programs.

One of my favorite quotes is “In God we trust. All others must bring data.” The best managers in any industry know how to transform the right data into invaluable information—and that information, in turn, into strategic initiatives.

Justin B. Bakst is managing director for Darling Consulting Group, a Newburyport, Mass.-based provider of asset liability management (ALM) solutions for financial institutions. 

Comments (0)