5 Questions to Optimize Risk and Return through a Deeper Understanding of Non-Maturity Deposits

Every bank across the country is required to conduct a non-maturity deposit study, but not all studies are created equal. These 5 questions will help you identify how effective your NMD study is and how you can ensure it’s working toward the goals of your bank.

Are your non-maturity deposit assumptions based on institution-specific analysis?

 Your customer base, product suite and overall franchise are unique. Thus, relying on industry estimates, averages or "what we have seen" consultant recommendations can result in risk profiles that are incomplete, or worse yet, incorrect. This occurs for multiple reasons including:

1. Underlying institutions are not comparable to your bank

2. The consolidation or averaging process tends to "mask" outliers in the underlying databases, even though those outliers may be the most reflective of your deposit/share portfolios

3. Providers of index or industry average data often apply additional adjustments to make results more "conservative" due to the more generic nature of the outputs

Using your own data to develop non-maturity deposit behavior assumptions, on the other hand, ensures that ALM model inputs are consistent with your specific pricing and customer/member behavior history.

How do you benchmark current NMD assumptions?

Once your NMD assumptions are based on reliable data, it is important to regularly review and assess their continued relevance.  Back-testing is one way to evaluate your current assumptions, however back-testing deposit balance retention or decay rates is difficult.

On the other hand, benchmarking current assumptions using an alternative methodology (like using your own data history), can provide additional support for assumptions. A benchmark analysis with comparable results provides additional assurance that your NMD assumptions are capturing and quantifying behavior forecasts appropriately. 

Is your NMD analysis solution cost-effective for the value of the results you receive?

As mentioned in an earlier post, industry averages, while generally inexpensive, can't always be relied on to provide accurate forecasts for your own NMD portfolio. Elaborate studies may offer limited value relative to their cost if methods are unnecessarily complex, the data collection requirements are too cumbersome or the results are difficult to implement. 

Numerous costs may come with an NMD study - direct costs such as consulting fees or indirect costs such as the resources needed to collect the necessary data – all of which can add up quickly. 

A cost-effective solution is achievable by leveraging existing data sources that offer a long history across multiple rate cycles and applying straightforward methods to calibrate the NMD beta and decay behavior assumption inputs needed for ALM modeling. 

Most importantly, it doesn't need to cost several thousand dollars!

Does your NMD analysis open a pathway to moving into a more detailed account study?

With your NMD study results, it’s crucial to be able to “walk your study forward” while also maintaining the ability to trace the expanded and granular category results to their more aggregated predecessors.

This approach helps document the new analysis and demonstrate sound governance over the development of updated NMD assumptions. For community institutions, in particular, it is important to follow a "Step~ Stretch~ Leap" path to build out NMD assumptions. 

Step: Begin with institution specific data at an aggregated level as indicated by budget and resource considerations

Stretch: As NMD assumptions are socialized and additional questions arise, move to a more detailed or comprehensive analysis that expands the NMD category set while maintaining reconcilability to the earlier aggregated NMD rollups.

Leap: As NMD behavior analytics move beyond regulatory compliance as a primary focus toward a more proactive balance sheet management orientation, expand data collection efforts to include additional depositor attributes

Regardless of whether you start or end with the Step~ Stretch~ Leap approach, it is important to ensure that each phase builds upon the earlier one.

How Often Do you Update Your NMD Assumptions?

Deposit portfolios, like the rest of the balance sheet are not static. Market and economic conditions change as do product pricing and strategic initiatives. Thus, it is critical to ensure that NMD behavior assumptions used for ALM modeling are up-to-date and consistent with current conditions.

However, many institutions grapple with the costs (internal and external) of conducting periodic refreshes of NMD analytics. Additionally, in most cases there is a lag (sometimes substantial) between the most recent data end date for the analysis and delivery of updated results. 

It is most advantageous to adopt a NMD modeling framework that is designed up-front to be regularly updated and with minimal delays in processing updated results. This ensures that your NMD forecasts are current and consistent with the most recent data and market conditions.

What’s Next?

After analyzing your own NMD studies through this lens, you may start to have more questions than answers. This is where LakeHouse Analytics comes in. Our NMD study software addresses each discussed question to ensure that you have every aspect of your NMD study covered. Our NMD study is often used in conjunction with current studies to fill in the pieces that other studies lack. Book a demo today and bring insights to your boardroom that everybody else overlooks.