Liquidity-Adjusted Duration: Incorporating Funding Risk into ALM Metrics

Liquidity-Adjusted Duration (LAD) is a modified duration metric that quantifies the sensitivity of a financial instrument’s or portfolio’s value to changes in both benchmark interest rates and funding costs-particularly unsecured term rates or overnight indexed swap (OIS) rates-while also reflecting the impact of liquidity risk on cash flow timing and reinvestment assumptions. Unlike standard effective duration, which assumes stable funding and liquidity conditions, LAD explicitly incorporates the second-order effect of funding rate shocks and liquidity stress on value, making it especially relevant for instruments with embedded options (e.g., prepayment, call, or early withdrawal features) or mismatched funding structures.

The adjustment arises because funding cost volatility can alter the effective maturity and reinvestment risk of assets and liabilities. For example, a rise in unsecured funding rates may increase the cost of rolling liabilities, thereby reducing the net present value of a fixed-rate loan portfolio funded with short-term liabilities-even if benchmark yields remain unchanged. LAD captures this by augmenting the standard duration formula with a funding sensitivity term, yielding a more realistic exposure measure for asset liability management (ALM) and interest rate risk measurement.

Liquidity-Adjusted Duration is derived by extending the first-order price sensitivity decomposition to include funding rate risk. For a portfolio value V, benchmark yield y (e.g., swap rate), and funding rate f (e.g., 3-month OIS), LAD is defined as:

where Δy and Δf represent parallel shifts in the yield and funding curves, respectively. The term ∂V/∂f is estimated using scenario analysis or regression-based sensitivity methods, often calibrated to historical funding stress periods (e.g., March 2023 banking turmoil) or stress test assumptions aligned with Basel III NSFR and LCR frameworks.

In practice, institutions compute LAD using dynamic cash flow models that simulate how funding cost changes affect liability rollover costs, deposit runoff behavior, and prepayment speeds. For instance, a rise in OIS rates may accelerate retail deposit outflows, shortening the effective duration of liabilities but increasing the funding gap for fixed-rate assets-LAD quantifies this feedback loop.

Liquidity-Adjusted Duration is increasingly embedded in internal ALM dashboards and regulatory reporting, particularly where funding risk and liquidity coverage ratios (LCR/NSFR) are monitored in tandem. According to ECB supervisory guidance, ALM frameworks must reflect the interplay between interest rate risk and liquidity risk, especially when funding sources are concentrated or wholesale-dependent. LAD serves as a bridge between these two risk types by translating funding volatility into a duration-like sensitivity metric.

Banks integrate LAD into:

• Stress testing: Simulating funding rate spikes (e.g., +200 bps in OIS) alongside yield curve shocks to assess capital and liquidity impact.
• NSFR alignment: Adjusting asset-side LAD to reflect the 1-year time horizon of the NSFR, where stable funding requirements vary by liquidity classification.
• Funding cost hedging: Using basis swaps or unsecured bond forwards to hedge ∂V/∂f exposure, complementing traditional duration hedging with benchmark swaps.

Consider a bank holding $1 billion in 30-year fixed-rate mortgages, funded with 3-month OIS-linked deposits. Standard effective duration might estimate a duration of 12 years. However, under a liquidity stress scenario where OIS rates rise by 150 bps:

• Deposit runoff accelerates, reducing funding stability.
• The bank must roll liabilities at higher rates, increasing net interest cost.
• Prepayment speeds decline as borrowers lock in lower rates, lengthening asset duration.

Liquidity-Adjusted Duration captures this combined effect: suppose ∂V/∂f yields an additional −2.5 years of sensitivity. The LAD becomes 12 + 2.5 = 14.5 years-indicating a 21% higher sensitivity to rate changes than standard duration suggests. This adjustment directly affects economic value of equity (EVE) and net interest income (NII) stress outcomes.

While LAD improves risk measurement fidelity, it introduces modeling complexity and data requirements:

• Model risk: Estimating ∂V/∂f requires robust behavioral models for deposits, prepayments, and wholesale funding, which are often noisy and regime-dependent.
• Calibration challenges: Funding rate sensitivity varies by instrument type, customer segment, and market structure-e.g., retail deposits show lower ∂V/∂f than wholesale funding, per Moody’s analysis of liquidity risk automation.
• Regulatory alignment: LAD is not yet standardized in Basel or OCC guidance; institutions must justify assumptions to supervisors, especially when LAD diverges significantly from standard duration.

Institutional adoption remains concentrated among larger banks with sophisticated ALM systems and exposure to volatile funding markets, particularly those relying on wholesale funding post-2023 banking sector stress. As liquidity risk monitoring evolves under the NSFR and ECB’s 2025 ALM framework, LAD is expected to gain broader use as a bridge between capital, liquidity, and interest rate risk metrics.