Jump to main content

Do We Need an ALM Validation?

September 2020
By Adam Stone, Vice President ALM & Investments, QuantyPhi

Credit unions have a lot on their plate right now. Over the last several months, credit unions have encountered an unexpected surge in deposits leading to rising asset levels and downward pressure on earnings and capital ratios. Furthermore, credit unions are facing a historically low interest rate environment, decreased loan demand, a spike in loan forbearance requests, double-digit unemployment, and overall economic uncertainty due to a pandemic.

Each of these issues poses unique constraints for a credit union to navigate. Foremost, credit unions must ensure they continue to serve their members, not only by providing loans and other financial services, but also through proper risk management of their institution’s operations and balance sheet.

Oftentimes, maintaining accurate and up-to-date asset and liability management (ALM) assumptions and performing an ongoing, thorough review of a credit union’s ALM process gets lost in the shuffle, serving as an afterthought or placed on the backburner while the aforementioned immediate, pressing concerns are handled. Each of these issues has the potential to drastically impact a credit union’s ALM risk profile.

Credit unions routinely alter their product offerings, offered rates, balance sheet composition, and business models. These changes have a domino like effect on a credit union’s risk profile, scope of operations, and earnings and capital levels. Whether these changes are through calculated management decisions or out of necessity due to economic conditions, it is important to ensure all data and assumptions feeding an ALM model are accurate and reflective of the current credit union position. It is also essential to verify the reasonableness of ALM model output and ensure the credit union is still in compliance with all applicable regulations and internal policies.

An integral part of ALM risk management is a periodic ALM validation performed by an independent, third-party. This is to ensure the end-to-end ALM process is accurate and best able to inform management decisions. Not only are ALM validations a regulatory requirement, they are central to maintaining safe and sound credit union operations.

All that said, credit unions may still ask themselves if they need an ALM validation. There are a myriad of reasons why a credit union may doubt the necessity or urgency of commissioning a validation. QuantyPhi is here to help.

If any of the following statements are true; then yes, your credit union needs, or would greatly benefit from, an independent, third-party ALM validation.

  • It has been two years or longer since our last ALM model validation. Regulators expect a credit union to have an ALM validation performed at least every two years. If it has been two years or more since your credit union has received an ALM validation, then it would be prudent to perform one now. Waiting until the regulators demand one could lead to unnecessary time and pressure constraints, not to mention increased regulatory scrutiny.
  • We have recently seen a large increase in member deposits. Most credit unions have seen a large increase in member deposits because of the pandemic and corresponding governmental stimulus activity. Regulators specifically require that management carefully consider deposit and non-maturity deposit (NMD) decay-rate assumptions, particularly when member behaviors change during periods of stress, as well as due to any external factors that may influence depository behavior. Recent events are a prime example of this. An ALM validation will carefully assess current and projected deposit levels and review assumptions relating to decay rates and market conditions. The recent influx of member deposits should not necessarily be treated the same as typical course-of-business deposits and may require a special set of assumptions.
  • Our ALM model vendor provides our ALM validation. Regulators expect every credit union to ensure their ALM model is appropriate for their interest rate risk profile by conducting an independent review and validation and performing ongoing monitoring and back-testing to confirm model appropriateness. Model certifications and validations provided by a model vendor can be useful but are not enough to satisfy the regulatory requirement. If you have relied solely on a model vendor validation, that is not meeting the regulatory requirement and an independent, third-party validation should be performed.
  • We recently switched ALM model vendors. ALM models can vary significantly depending on complexity, data management, strategies, model-provided assumptions and market rates, and their ability to sufficiently capture risks of a credit union. Furthermore, the process to provide data and assumptions to the model provider, and the way the provider utilizes the data and assumptions are unique. Following a switch of ALM model providers, a credit union should work to optimize the process over two to three quarters of model runs, and then perform an ALM validation to ensure the transition was smooth and the new vendor is accurately reflecting the risk position of the credit union.
  • There have been changes to our product offerings, business model, or risk profile. Any new type of product offering should be thoroughly vetted prior to launch, including an analysis of any effect to the credit union’s ALM position. ALM assumptions need to be developed and assigned to the product and incorporated into the ALM model. Larger scale changes to business models or operations also need to be implemented with a focus on ALM risk management and communicated to the ALM model vendor. An ALM validation will ensure the model is accurately capturing these changes, which can at times be difficult to implement and track.
  • We made changes to our interest rate risk (IRR) and/or ALM policy. Credit unions should continually review the effectiveness of IRR and ALM policies to ensure they reflect the appropriate measurement methodologies, for example Net-Interest Income (NII), Net Income (NI), and Net Economic Value (NEV). Policies should further be reviewed to ensure the appropriateness of scenarios (e.g. +/- 100 bps, +/- 300 bps), and any policy limits or warning indicators in accordance with the credit union’s risk tolerance. Any changes to an IRR or ALM policy need to be implemented in the ALM model, and an ALM validation will confirm the changes are reflected correctly.
  • We rely on industry estimates or default vendor assumptions for NMD decay rates and other key assumptions. Credit unions should utilize assumptions that reflect their own risk profile and activities and generally avoid reliance on industry estimates or default vendor assumptions. Granted, some credit unions have system limitations or lack the historical data required for a proper NMD decay rate analysis, and industry averages or ALM vendor provided assumptions can help to fill in the gaps. However, whenever possible, assumptions should be carefully tailored to the unique position of each credit union. An ALM validation can determine the accuracy and usefulness of current assumptions across product categories, and upon request, include an NMD decay rate analysis to further refine deposit assumptions.
  • We do not routinely review and update our ALM assumptions. As markets change, so should future-looking assumptions. Proper management and measurement of IRR requires credit unions to regularly assess the reasonableness of all ALM assumptions. Credit unions should document, monitor, and regularly update key assumptions used in IRR measurement models. This includes the reasonableness of asset prepayments, NMD decay rates, price sensitivities, and key rate drivers for each interest rate shock scenario. An ALM validation will closely look at all assumptions and test for reasonableness in respect to current credit union activities and economic conditions. The ALM validation results will include any recommended assumption changes to reflect the credit union’s balance sheet more accurately.
  • We only measure gap risk and/or re-pricing risk. All credit unions are expected to run stress scenarios commensurate with their risk profile on a periodic basis. This allows a credit union to identify all significant risks attributable to the main components of IRR, including repricing mismatch, basis risk, yield curve risk, and options risk. Credit unions should conduct these additional stress tests on at least an annual basis, and if a source of IRR is discovered, that particular stress test should be incorporated into routine quarterly model runs. An ALM validation can help determine if the appropriate stress scenarios are performed, as well as interpret the results and suggest the appropriateness for inclusion in routine monitoring.
  • We have received negative regulator feedback surrounding our ALM process. During their review, regulators may identify specific areas in your credit union’s ALM process as being deficient. An ALM validation will not only review the entire end-to-end process but can also focus on certain areas and guide a credit union through their regulatory response to ensure all problem areas are remediated.
  • We do not fully understand the output of our ALM model and how to interpret the results. Regulators identify a credit union’s board of directors as having the ultimate responsibility for the risks undertaken by an institution, including IRR. The board should be regularly informed and educated on their credit union’s ALM position and any significant internal or external factors that play a role. Senior management, including an asset-liability committee (ALCO), is responsible for ensuring that board-approved strategies, policies, and procedures for managing IRR are appropriately executed. It is imperative that senior management and the board of directors understand the assumptions feeding an ALM model, as well as the model output. An ALM validation will review the effectiveness of ALM governance and clearly articulate any problem areas. Furthermore, upon request, the ALM validation can be accompanied with board and/or senior management training.
  • Our ALM vendor recommends using a specific provider for validations. The very nature of an independent, third party validation relies on the credit union choosing an ALM validation provider that is unbiased and impartial to any specific ALM model vendor, or any other potential conflicts of interest. It is important that a credit union independently choose a validation service that understands their credit union balance sheet while also being free of any possible quid-pro-quo arrangements with a model vendor. Validations start with a clean slate and look at the ALM process from end-to-end, including both credit union provided data and assumptions as well as any assumptions the model itself provides and how the model utilizes the data and delivers output. It is wise to perform your own research on validation services and choose the one that best understands your balance sheet and can provide the best fit service for your needs.

How can QuantyPhi help?

QuantyPhi’s goal is to help your credit union evaluate the interest rate risk on your balance sheet and provide insight on how to improve your credit union’s performance and pass your exams with ease and confidence.

While conducting your credit unions ALM validation, QuantyPhi’s asset liability experts will review critical areas of IRR measurement and management including:

  • Data input and balancing procedures
  • Determination and reasonableness of assumptions
  • Accuracy of reporting
  • Compliance with regulations and policies
  • ALCO and board reporting
  • Governance and separation of duties

By using powerful state-of-the-art technology, our highly-trained analysts will review your credit union’s IRR process, pinpoint areas that are not currently optimized for overall peak performance, suggest methods and practices to help improve your credit union’s asset liability management process, and help you implement any desired changes.