Interest Rate Risk (IRR) analysis and measurements are commonly obtained from traditional in-house asset liability management (ALM) models, outsourced services or vended models. There are three general ALM model levels: entry, intermediate and advanced.

Entry-level ALM models are the simplest of ALM modeling solutions. They provide basic levels of net interest margin simulation for a limited shock outlay. The problem is that entry-level ALM models often cannot adequately address option-related complexities, such as prepayments embedded in mortgage-backed fixed rate loans or securities. They also generally have limitations with regards to behavioral modelling and classification of non-term deposits. Entry-level models make use of hard-coded assumptions, such as a fixed prepayment rate and use of the laddered approach to determine the term-to-maturity of non-term deposits. This impacts the net repricing gap, cash flow hedging and limits the liquidity survival horizon.

Intermediate-level ALM models utilize rigorous statistical algorithm to hedge basis and repricing risks. Equity immunization is achieved by setting DA – k*DL = 0, where, DA=Duration of Assets, and DL=Duration of Liabilities. In other words, the difference between duration of assets and liabilities adjusted for factor “k” must equal zero to immunize equity. Whereas, a fixed prepayment rate in Entry-level model creates a lopsided distribution and causes the run-in balances to shift unwarrantedly. Intermediate-level models use a statistically-driven prepayment model that acts as a function of past prepayment rates and expected future economic conditions to smoothen the distribution. Non-term deposits are first classified into core, volatile and rate-sensitive cohorts; duration is then assigned, which helps create a reliable pool of liquidity to support longer term transactions and accuracy in interest rate sensitivity. Intermediate ALM models should also be able to create scenarios (such as yield curve shape change) and test the impact of rate shocks based on such scenarios. Another feature of intermediate ALM models is customized dashboards and automated ALM reporting. An added benefit of the Intermediate ALM model is that many models can also simultaneously interface with budgeting and planning applications.

A weakness of intermediate models, however, is their inability for record level processing, such as treasury transactions.

Intermediate-level ALM models are the right choice for institutions that have needs beyond entry-level models but do not require an ALM model with the most advanced capabilities or have a separate Treasury Management System.

The definition of an “advanced” ALM model has broadened over the last several years. Advanced-level ALM models are the solution for institutions that need maximum ALM model capabilities. This may be due to balance sheet complexity, derivative instrument valuations, OTC transactions, special business lines or regulatory mandates.

BBA’s Asset Liability Management solution is a part of BB-ALMTM. BB-ALM is BBA's proprietary financial and enterprise risk management suite of applications developed for regulated financial entities. It can be classified as an Intermediate ALM model.

BBA's ALM solution is organized in three parts: Static ALM, Dynamic ALM and ALM Stress Testing. Static ALM organizes on and off balance sheet volumes of assets and liabilities in discrete time buckets based on their run-ins, i.e., principal repayments and prepayments. It also generates two key metrics, discrete and cumulative repricing and liquidity gaps.

The second part is dynamic ALM. Dynamic ALM shock tests both parallel and non-parallel rate movements. Dynamic ALM has two parts: Market Value at Risk and Net Interest Income at Risk. Market Value at Risk is determined for each line item on the balance sheet. First, the application develops principal gapping tables, then the duration of on and off balance sheet assets, liabilities and equity, and finally maps them to key rate nodes on the yield curve and then shocks each node to determine Market Value at Risk. The key objective of this metric is to immunize equity to protect capital.

For Net Interest Income at Risk, BBA's ALM solution automatically shock tests repricing volumes to determine the worst case NIM. For the worst case shock, we investigate the kind of hedges required for basis and cash flow risks.

The third part is the stress testing of Market Value at Risk and Net Interest Income at Risk by ramping up and down the rates. From a plausibility perspective we determine the implied volatility of market rates relative to adverse shocks and identify emerging hedging needs and management’s action plan.

BBA's ALM solution has the following unique value propositions:

• Designing interest rate scenarios based on projected implied volatility of rates - not fixed shock tests
• Accuracy of calculations - our calculations are third party reviewed and confirmed
• Safeguarding income and capital by using our complimentary cash flow hedging and equity immunization for efficient capital management

Finally, the model also extends to Residential Mortgage Backed Securitization. In most cases our clients are not required to nominal capital for Interest Rate Risk in the Banking Book. Our ALM model is suitable for banks holding less than $40.0 Bn in assets. BBA's ALM solution offers three use cases:

• The first and foremost is for banks still using spreadsheets to manage interest rate risk. The broader the use of spreadsheets, the greater the chance for errors to be magnified, exposing the organization to significant costly risks. Our customized ALM solution can also be integrated with the Treasury Management System and feeds into budgeting and planning applications. This also allows for near real time liquidity management. Other features include automated ALCO reporting, data management, archiving and the development of customized dashboard.
• The second use case is for those clients who consider they are paying too much for their ALM solution or are at an entry-level model and wish to transition to a more sophisticated version.
• The third use case is for banks looking into getting their ALM systems certified. There are two approaches to our ALM models validation: i) A full replication of the model being used; and ii) A rigorous testing of the assumptions, calculations and methodology. We help our clients create full confidence in their model and its ability to produce accurate, reasonable and stable results.

For banks less than $5 Bn in assets, we are offering a promotion on ALM modules' subscription license. This includes no implementation fee and a free 60 day trial. In other words this is a risk-free offer. For model certification and testing we are extending a special offer of USD/CADEq $13K/model with results delivered in 5 business days. Terms apply.

BBA’s ALM solutions video demo can be seen here.

For more information on this topic, or to learn how the Team BBA can assist your organization, contact our team.