Structural Liquidity Planning and Optimal Hedging Portfolios

Book: Financial Risk Management: Applications in Market, Credit, Asset and Liability Management and Firmwide Risk Authors: Jimmy Skoglund & Wei Chen ISBN: 978-1-119-13551-7

Hedging liquidity in a crisis is reactive. Structural planning is proactive: build the right buffer and balance sheet before stress arrives.

Two complementary approaches

Structural balance sheet management: acquire assets and liabilities with naturally offsetting cash flows. Long-term deposits funding long-term loans. Commitment facilities priced to reflect draw risk.

Contractual cash flow hedging: hold a portfolio of liquid assets whose maturity profile matches expected stressed outflows.

Banks need both. Natural hedges rarely cover extreme stress because of embedded options (deposit withdrawals, facility draws, prepayments).

Cash flow matching optimization

Like bond portfolio immunization, but for liquidity. Find the minimum-cost asset portfolio that generates enough cash at each future date to cover stressed net outflows.

Chen and Skoglund (2012) use linear programming with CVaR constraints on the risk reserve process. You choose asset holdings to minimize upfront cost while keeping the liquidity buffer above tolerance levels at each horizon.

The model can run with zero initial counterbalancing capacity, forcing the optimizer to find assets that inherently cover the gap. Or it can layer on top of existing buffers.

Balance sheet vulnerability

Structural planning also asks whether the balance sheet itself is the problem. Too much short-term wholesale funding? Too many committed lines with low fees that encourage drawdown? Maturity mismatch between assets and liabilities?

Acquiring complementing flows (e.g., more sticky retail deposits) reduces the buffer size needed. That is cheaper than holding excess cash forever.

Opportunity cost tradeoff

HQLA earns less than lending. The bank wants the smallest buffer that survives stress scenarios with acceptable probability. Minimize opportunity cost subject to survival constraints.

This connects directly to funds transfer pricing in Chapter 7: the cost of holding standby liquidity should flow into product pricing.

Dynamic capacity

Repo markets, asset sales, and credit facilities provide dynamic counterbalancing capacity conditional on current balance sheet and market conditions. Structural planning sets the static buffer; execution models handle the dynamic part.

Natural hedging limits

Diamond-Dybvig style bank runs show why retail deposit franchises have value beyond funding cost. Kashyap, Rajan, and Stein analyze how combined deposit and lending businesses create natural liquidity hedges. But embedded options break simple matching. Callable deposits and committed credit lines are options sold cheaply until stress reveals the true cost.

Structural planning asks whether to originate different products (more sticky liabilities, shorter asset duration) rather than only hold more HQLA.

Immunization analogy

Bond immunization matches duration and convexity. Liquidity immunization matches cash flow timing. The LP model in Chen and Skoglund minimizes upfront cost of a bond portfolio subject to CVaR constraints on the liquidity reserve at each future date. Negative lambda thresholds allow controlled mismatch when you already hold a separate buffer.

My take

This section is where liquidity risk stops being a treasury ops problem and becomes a balance sheet design problem. The LP optimization is technical, but the intuition is simple: match your rainy-day fund to your rainy-day outflows as cheaply as possible.

Banks that only react in crisis overpay for liquidity every year. Banks that plan structurally hold less dead cash and still survive stress.

The link to FTP in the next chapter is direct: structural liquidity cost should appear in transfer prices. Otherwise branches keep writing long-dated assets funded by assumptions of perpetual cheap deposits.

Treasury should revisit the structural plan when product mix shifts. A mortgage boom or a push into wholesale funding changes the stressed outflow curve. Static buffers go stale quietly until the next stress event proves it.

Previous: Hedging Liquidity Exposure Next: Liquidity Measures and Regulation