How Banks Measure Liquidity Risk with Cash Flow Models

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

Chapter 6 shifts from derivatives desks to the balance sheet. Liquidity risk is the risk you cannot pay what you owe when it is due, without taking a big loss. Skoglund and Chen treat it as a first-class quantitative problem, not just a regulatory checkbox.

Why liquidity is different

Market and credit risk get capital buffers. Liquidity risk gets cash. The hedge for liquidity is ultimately liquid assets you can sell or pledge, not more equity.

A bank can be long-term solvent but short-term illiquid. Profitability does not guarantee you can roll over funding tomorrow.

The 2007 crisis as a liquidity story

Subprime credit losses triggered the crisis, but the killing blow was liquidity. Interbank lending froze. LIBOR-OIS spreads went from single digits to 350 basis points. Northern Rock could not roll short-term money market funding. Bear Stearns, Lehman, AIG: funding problems, not just bad assets.

Liquidity risk is usually consequential. Credit losses or market shocks hit first. Then counterparties demand more margin, depositors run, and funding markets close. Reputation drives everything.

Traditional measures

Banks have long used funding gaps: cash inflows minus outflows over time buckets, under normal and stressed assumptions. Liquid assets hedge the gap. Structural ratios like liquid assets over non-sticky deposits give a quick health check.

A ruin-theory approach

Unlike VaR for market risk, liquidity solvency is path-dependent. You can survive at t+1 but fail at t if you cannot convert hedging assets to cash fast enough.

Skoglund and Chen adapt insurance ruin theory. Build a risk reserve process from cumulative cash flows plus hedging capacity. Insolvency is the first time that process goes negative.

Under stress scenarios, you project net cash flows at each time bucket. Positive flows roll forward. Negative flows must be covered by the liquidity buffer.

Cash hedging capacity simplifies things

If the hedging portfolio is cash or cash equivalents (with haircuts), liquidity measurement becomes monotone over time. You can define solvency probability cleanly, similar to classical risk measures.

This aligns with Basel III’s liquidity buffer definition: eligible assets converted to cash equivalents using regulatory haircuts.

Components of the liquidity measure

The full measure needs:

• Stressed cash flow projections (behavioral assumptions on deposits, loan prepayments, facility drawdowns)
• Counterbalancing capacity from the hedging portfolio (sale, repo, or draw on credit lines)
• Time buckets matching the planning horizon

Liquidity exposure with general hedging capacity is more complex because asset liquidation is uncertain. Haircuts, fire-sale discounts, and market access all matter.

Risk reserve process in plain terms

Picture a running cash balance that starts with your liquidity buffer. Each day, stressed net outflows drain it. Asset sales and repo proceeds refill it, minus haircuts. The first date the balance hits zero is your insolvency date for that scenario.

Limits are set on the minimum reserve level at each horizon. Treasury monitors the process like market risk monitors VaR breaches. The difference is timing: liquidity failures are discrete cash events, not smooth P&L swings.

My take

The path-dependency point is the key insight. You cannot just take a 30-day VaR mindset and apply it to liquidity. The first day you miss a payment, you are done, even if next week’s inflows would have saved you.

The crisis narrative in this chapter is worth reading even if you skip the formulas. It explains why Basel III added hard liquidity ratios after capital rules failed to prevent runs.

Skoglund and Chen borrow ruin theory because it forces you to think about survival probability, not just average funding need. That mindset shift is what separates ALM teams that survived 2008 from those that did not.

Previous: Counterparty Credit Risk Regulation Next: Liquidity Exposure Analysis