Asset Liability Management

Life insurance actuaries have developed time-tested liability models, reinforced by experience that forms the core of the regulatory cash flow test. These models, stressed by dynamic interest rate scenarios, have for years provided regulators and management with a profile of the behavior of the business unit:

• If interest rates rise, can the investment portfolio support expected lapse?
• If interest rates fall, can the investment portfolio continue to pay claims and minimum guarantee rates?
• ” Does the business unit have an “Achilles Heel” – a circumstance where an asset liability match fails

Successful companies have evaluated these results to identify a asset liability match that best immunizes the business from adverse results – losses – or worse, insolvency.

It was this understanding of dynamic modeling that caused AQS to develop the Precision Asset Liability Management (PALM) software. Using the very same liability inputs that produce annual testing results, AQS identifies a singularly superior portfolio structure – optimized across a range of interest rate scenarios – that immunizes the business unit against adverse rate environments.

Duration matching falls short of this goal. Though sound in concept, the duration match presumes marketability of assets or liabilities in the future. Market professionals need look no farther than 2009 when credit spreads ballooned and liquidity was nonexistent. The market for liabilities is even less liquid.

It is this flaw in the duration match that caused AQS to consider a platform that could exploit the confluence of new technology and information delivery.

The PALM system:

• Produces a portfolio blueprint more robust in financial performance with a level of certainty trial-and-error modeling cannot replicate.
• Provides granular (specific as to CUSIP) and tractable portfolio solution within a market-critical timeframe.
• Adds certainty to the ALM model allowing AQS portfolios to be less exposed to event-driven credit risk which cannot be modeled or predicted.