Recommendations for a Successful ALM Risk Management Regime

Category: Data Optimization
Author: NXTsoft
 The days of going through the basic motions on Asset Liability Management are over. Regardless of assets size, balance sheet complexity, charter type and regulatory body, examiners are expecting more and more from your ALM process. Even small institutions with simple balance sheets are being asked to do much more with respect to their ALM process. It’s clearly time to up your game. 

The most common types of regulatory requests we see include the following: 

• Stress Testing of Key Modelling Assumptions 
      o Non-maturity account decay (or lives) 
      o Product pricing sensitivities (Betas) 
      o Loan prepayment rates 

• Establishment of key modelling assumptions based on your own historical customer behaviors 
      o Non-maturity account decay (or lives) 
      o Product pricing sensitivities (Betas) 
      o Loan prepayment rates 

• Alternate Rate Scenarios 
      o Non-parallel rate changes 
      o Ramps 

• Enhanced back testing of your ALM results (short and longer termed back testing periods) 

• Liquidity scenario modelling 
      o Regular liquidity ratios and projections 
      o Testing of contingency liquidity scenarios 

In addition to all that, you are expected to have policies and procedures built around all of these. Then when exam time comes, you need to be prepared to explain, discuss and defend all that you have done. And it’s not just senior management and ALCO that need to be competent in these areas. Your board has responsibilities to be kept informed and understanding of these matters as well. 

The days of waiting to prep for your next exam a few weeks before it starts are long gone. You are well advised to create systematic process to deal with small portions of these each quarter between your examinations. This approach will also make it much easier to keep your ALCO and Board informed on these matters without risking sensory overload. 

The good news is, with the exception of liquidity modelling, most institutions can get away with a once a year to once every 18 months for running these scenarios. Unless you have an incredibly dynamic balance sheet, running these alternate scenarios will produce similar impacts each time they are run, so less frequent processing should be sufficient. 
 


Create a calendar for your ALM modelling program, taking on one or two additional tasks each quarter throughout the year, and stick to it. 

 When it comes to updating your key modeling assumptions, run your model with your old assumptions first, then run them again with the new assumptions. This way the reader of the report can differentiate the changes in your results compared to the previous reporting period due to the changes in your balance sheet etc., and then compare that run to the run with updated assumptions in order to assess the impact of the new assumptions separately. 

 Examiners have for a long time expected you to review and update your key modelling assumptions regularly. Normally, a once a year review has been commonplace. More frequent updates would be appropriate for any of the following reasons: 
• Change in business lines or strategies, 
• Entry of a new or more aggressive competitor which impacts you or your customers behaviors, 
• Acquisition or merger that takes you into new markets, 
• A significant change in the external rate environment, or 
• Any other event, such as a natural disaster impacting you or your customers behaviors. 

 For years your examiner has probably asked you to base your key modelling assumptions around results of historical behavioral studies. By now they probably really mean it. Whether you conduct these studies in-house, or out-source them, you will likely have fair bit of work to do just in collecting the data that is required. We will provide additional insight on these studies in future blog posts, but don’t let that stop you from beginning the basic data collection process. 

 Keeping ahead of your regulator will help ensure a positive exam outcome and these steps should help you get there. 

 NXTSoft offers a wide range of risk management products and services. Please submit requests for additional information to info@nxtsoft.com or call 877-322-1800 x706. 
July 17, 2019
Back
Share this post on social media