Liquidity Risk and Asset–Liability Management

Introduction

Banks transform short-term, withdrawable funding into longer-term loans and investments. This maturity transformation supports economic activity but creates liquidity and interest-rate risk. A solvent bank may fail if it cannot obtain cash when obligations fall due, while changes in market rates can materially alter net interest income and the economic value of equity.

Liquidity management therefore concerns more than holding cash. It integrates cash-flow forecasting, depositor behavior, collateral management, funding diversification, stress testing and contingency planning. Asset–liability management (ALM) coordinates these activities with the management of interest-rate risk in the banking book (IRRBB).

Learning Objectives

After studying this chapter, students should be able to:

1. Distinguish funding liquidity from market liquidity.
2. Construct and interpret cash-flow gap and maturity-mismatch schedules.
3. Evaluate deposit stability, deposit beta and withdrawal behavior.
4. Calculate and interpret major liquidity ratios.
5. Design liquidity stress tests and contingency funding plans.
6. Apply repricing-gap, duration and economic-value techniques to IRRBB.
7. Explain ALM governance, limitations and contemporary practices.

1. The Nature of Liquidity Risk

1.1 Funding liquidity risk

Funding liquidity risk is the risk that an institution cannot meet payment obligations when due without incurring unacceptable losses. Obligations include deposit withdrawals, wholesale debt maturities, margin calls, committed credit-line drawdowns and settlement payments.

Funding liquidity has two related dimensions:

• Stock liquidity: immediately available cash and readily monetizable assets.
• Flow liquidity: expected and stressed cash inflows and outflows over time.

A bank can possess valuable long-term assets yet experience a liquidity crisis if those assets cannot be sold or pledged quickly enough.

1.2 Market liquidity risk

Market liquidity risk is the risk that an asset cannot be sold promptly in sufficient volume without a substantial price concession. It is reflected in widening bid–ask spreads, falling market depth, longer execution times and increased price impact.

Funding and market liquidity reinforce each other. A bank facing withdrawals may sell securities. Forced sales depress prices, reduce collateral values and generate further margin calls. This feedback can transform an institution-specific problem into system-wide stress.

Figure: Adverse interaction between funding liquidity and market liquidity.

1.3 Liquidity risk versus solvency risk

Solvency means that the economic value of assets exceeds liabilities. Liquidity means having cash or reliable funding when payments are due. The distinction is analytically useful, but crises connect them: liquidity losses can force asset sales below fundamental value, erode capital and create insolvency concerns.

2. Cash-Flow Gaps and Maturity Mismatches

2.1 Constructing a maturity ladder

A maturity ladder assigns expected cash flows to time buckets such as overnight, 2–7 days, 8–30 days, 31–90 days and beyond. Banks should include:

• contractual loan and security receipts;
• deposit withdrawals and debt maturities;
• interest, operating and settlement payments;
• committed facility drawdowns;
• derivative cash flows and collateral calls;
• realistic asset-sale proceeds after haircuts; and
• behavioral assumptions for items without contractual maturity.

For bucket t, the net liquidity gap is:

where G_t is the net gap in currency units, I_t is cash inflows and O_t is cash outflows during bucket t. A negative value represents a shortfall.

The cumulative gap through bucket T is:

This arithmetic assumes that unused positive cash is carried forward. Actual usability may be constrained by currency, legal entity, collateral location or transfer restrictions.

2.2 Worked example: contractual liquidity ladder

A bank forecasts the following cash flows, in millions of currency units:

The maximum cumulative deficit is 60 million overnight. If the bank has 55 million of immediately usable counterbalancing capacity, it still has a 5 million overnight shortfall. A positive cumulative position after 90 days does not solve the earlier timing problem.

2.3 Counterbalancing capacity and survival horizon

Counterbalancing capacity consists of cash and assets that can realistically be monetized or pledged under stress. It should be measured after operational delays, market-value changes, regulatory haircuts and existing encumbrance.

If C_T is cumulative usable counterbalancing capacity by time T, remaining liquidity is:

The survival horizon is the last point before R_T becomes negative under a specified stress scenario. It is scenario-dependent rather than an intrinsic characteristic of the bank.

3. Deposit Behavior and Funding Stability

3.1 Contractual and behavioral maturity

Non-maturity deposits—such as current and savings accounts—may be withdrawable immediately but often remain for years. Consequently, ALM uses behavioral rather than purely contractual assumptions. Deposits may be segmented by customer type, account purpose, balance concentration, insurance coverage, rate sensitivity and digital withdrawal capability.

Typical segments include:

• operational retail deposits;
• non-operational corporate deposits;
• insured and uninsured balances;
• rate-sensitive wealth deposits; and
• public-sector or financial-institution deposits.

Historical stability must not be assumed to continue during reputational distress. Concentrated, uninsured and digitally mobile deposits may run much faster than diversified insured retail balances.

3.2 Deposit beta

Deposit beta measures the pass-through of a market-rate change to the rate paid on deposits:

where β_D is dimensionless, Δr_D is the change in the deposit rate and Δr_M is the change in a relevant market rate, both measured in the same rate units. For example, if the market rate rises by 2 percentage points and a deposit rate rises by 0.8 percentage points, beta is 0.8 ÷ 2.0 = 0.40.

Beta can differ in rising and falling rate environments. It also changes with competition, customer segment and the speed of rate adjustment.

3.3 Deposit decay

Under a simplified constant attrition assumption, the surviving balance is:

where B₀ is the initial balance, a is the attrition rate per period, and t is the number of periods. This model is a special case, not a universal law. Stress conditions usually require scenario-specific runoff rates rather than extrapolation from normal periods.

4. Liquidity Ratios and Market Indicators

4.1 Liquidity Coverage Ratio

The Liquidity Coverage Ratio (LCR) evaluates short-term resilience:

High-quality liquid assets (HQLA) are included after applicable eligibility rules and haircuts. Under the Basel framework, net outflows are generally:

The inflow cap prevents excessive reliance on expected receipts. Jurisdictional implementation may differ. A minimum of 100% generally means that eligible HQLA cover modeled 30-day net outflows under regulatory stress.

Example. HQLA after haircuts equals 150 million, outflows equal 220 million and eligible inflows equal 80 million. The inflow cap is 0.75 × 220 = 165 million, so all 80 million is recognized. Net outflows are 140 million and:

The bank exceeds 100%, but the ratio does not capture every institution-specific or longer-term vulnerability.

4.2 Net Stable Funding Ratio

The Net Stable Funding Ratio (NSFR) evaluates structural funding over approximately one year:

Available stable funding applies regulatory stability factors to capital and liabilities. Required stable funding applies factors to assets and off-balance-sheet exposures according to liquidity characteristics and maturity. An NSFR of at least 100% indicates that weighted stable funding is no less than weighted funding requirements.

4.3 Supplementary indicators

No single ratio is sufficient. Effective monitoring combines regulatory metrics, internal limits, cash-flow forecasts and market intelligence.

5. Liquidity Stress Testing

Stress testing asks whether the bank can survive severe but plausible conditions. Common scenarios are:

1. Idiosyncratic stress: institution-specific downgrade, fraud, cyberattack or reputational event.
2. Market-wide stress: disruption of wholesale markets, collateral-price declines or systemic deposit movement.
3. Combined stress: institution-specific weakness during general market turmoil.

Assumptions may include accelerated deposit runoff, non-renewal of wholesale funding, committed-line drawdowns, larger collateral calls, reduced asset-sale prices and loss of secured borrowing capacity.

Good stress tests are granular by currency, legal entity and time bucket. Reverse stress testing begins with failure—such as exhaustion of usable liquidity—and identifies the assumptions or events capable of producing that outcome.

Model limitations are significant. Historical data may not represent a digitally accelerated bank run; correlations may rise during stress; and supposedly liquid assets may become difficult to sell. Banks should therefore use multiple scenarios and managerial overlays rather than one mechanically estimated forecast.

6. Contingency Funding Plans

A contingency funding plan (CFP) specifies how an institution will identify and respond to liquidity stress. It should contain:

• quantitative and qualitative early-warning indicators;
• escalation thresholds and decision authority;
• a crisis-management team and communication protocol;
• a prioritized inventory of funding sources;
• collateral location and mobilization procedures;
• operational tests of central-bank and secured facilities;
• legal-entity and currency restrictions; and
• post-event review procedures.

Potential actions include monetizing unencumbered securities, increasing secured borrowing, drawing committed facilities, issuing term debt, slowing new lending and adjusting deposit pricing. Asset fire sales and emergency pricing should not be treated as costless.

A credible CFP distinguishes availability from usability. A nominal credit line is not reliable if it contains adverse-change clauses, depends on the same stressed counterparties or cannot be operationally drawn quickly.

Communication must be truthful and coordinated. Concealing material liquidity problems, misleading depositors or manipulating classifications is unethical and may intensify systemic harm.

7. Asset–Liability Management

ALM is the coordinated management of balance-sheet composition, liquidity, funding, interest-rate exposure and capital effects. The Asset–Liability Committee (ALCO) commonly approves limits, assumptions, hedging actions, funding strategy and product pricing.

A sound ALM process follows a cycle of data collection, measurement, stress testing, limit comparison, management action and independent validation. Treasury usually executes transactions, risk management provides independent oversight, finance reconciles accounting data, and internal audit evaluates governance and controls.

7.1 Funds-transfer pricing

Funds-transfer pricing (FTP) assigns internal funding and liquidity costs to business units. A well-designed FTP rate may include a maturity-matched benchmark, term-liquidity premium, contingent-liquidity cost and product-optionality adjustment. There is no single universal FTP formula; design should reflect the institution’s funding structure.

FTP discourages business units from reporting apparently profitable long-term loans without recognizing the cost of stable funding. Poor FTP can reward excessive maturity transformation or unstable deposit gathering.

8. Interest-Rate Risk in the Banking Book

IRRBB arises when banking-book assets, liabilities and off-balance-sheet positions respond differently to interest-rate changes. Major forms include:

• repricing risk: timing differences in rate resets;
• yield-curve risk: nonparallel movements across maturities;
• basis risk: imperfect movement among reference rates;
• option risk: customer behavior embedded in prepayments, withdrawals or caps; and
• credit-spread effects: valuation changes not fully explained by risk-free rates, where included by the applicable framework.

Two complementary perspectives are used: earnings-based measures and economic-value measures.

8.1 Repricing gap and net interest income

For a defined time bucket:

where RSA is rate-sensitive assets and RSL is rate-sensitive liabilities, both in currency units. Under a simple parallel, immediate and equal rate change:

Here ΔNII is the approximate annual change in net interest income and Δr is the rate change in decimal form. The formula assumes full-period exposure, equal pass-through and no balance-sheet growth, optionality or basis differences.

Example. If RSA is 420 million and RSL is 510 million, the gap is −90 million. For a 1.5 percentage-point increase, Δr = 0.015:

Actual results may differ because deposit beta, repricing lags and customer behavior violate the simplifying assumptions.

8.2 Economic value of equity

Economic value of equity (EVE) is the present value of asset cash flows minus the present value of liability cash flows:

Using discount factors DF_k appropriate to maturity and risk:

where CF_k is the cash flow at date k. EVE captures longer-term value sensitivity that may not appear immediately in one-year earnings.

8.3 Macaulay, modified and effective duration

For fixed cash flows discounted at an annual effective yield y, Macaulay duration is:

where P is present value and t is measured in years. Macaulay duration has units of years.

Modified duration under annual compounding is:

For a nominal annual yield compounded m times, the conversion is D_mod=D_Mac ÷ (1+y_nom/m). Modified duration gives the first-order price sensitivity:

No additional division by (1+y) is applied when duration is already modified.

For instruments with rate-dependent cash flows, such as callable bonds or mortgage loans, effective duration is generally more appropriate:

where P_- is value after a downward yield shock, P_+ is value after an equal upward shock, P₀ is current value and Δy is the absolute shock in decimal form. Effective duration already measures price sensitivity and must not be divided again by (1+y).

8.4 Duration-gap approximation

If asset and liability modified durations are D_A and D_L, respectively, and both are exposed to the same small parallel yield change:

where A and L are asset and liability present values. Equivalently:

These formulas use modified or effective durations, not unadjusted Macaulay durations. They are first-order approximations and become less reliable for large or nonparallel shocks, embedded options and substantial convexity.

8.5 Worked EVE calculation

Suppose a bank has:

• assets of 1,000 million with modified duration 3.8;
• liabilities of 900 million with modified duration 1.6; and
• a parallel increase in yields of 100 basis points, so Δy=0.01.

Asset value changes by:

Liability value changes by:

Because EVE=A-L:

Initial EVE is 100 million, so approximate post-shock EVE is 76.4 million. The calculation correctly uses modified duration without an additional yield adjustment.

Convexity can improve the approximation:

where C=(1/P)(∂²P/∂y²) under the stated yield convention. Convexity is most relevant for large shocks or instruments with curved price–yield relationships.

9. Illustrative Case: Meridian Commercial Bank

Background

Meridian Commercial Bank is a fictional mid-sized bank with 2,000 million in assets, 1,840 million in liabilities and accounting equity of 160 million. It has expanded fixed-rate lending using short-term corporate deposits.

Problem

Under the baseline regulatory calculation, HQLA is 210 million, 30-day outflows are 300 million and eligible inflows are 100 million. Net outflows are 200 million, producing an LCR of 105%. Management initially regards this as comfortable.

A combined stress changes the position:

• HQLA falls to 195 million after market-price changes and haircuts;
• outflows rise to 360 million;
• eligible inflows fall to 80 million; and
• the cumulative 30-day cash-flow deficit reaches 225 million.

Analysis

The stressed inflow cap is 75% × 360 = 270 million, so all 80 million of inflows is recognized. Therefore:

The stressed liquidity buffer is 30 million below the 225 million cumulative deficit. Meridian cannot survive the full 30-day scenario without additional action.

Its one-year RSA is 720 million and RSL is 900 million. For a 2 percentage-point rate increase:

Asset modified duration is 3.6 and liability modified duration is 1.4. For the same 2% parallel shock:

The calculation indicates substantial long-term value sensitivity as well as short-term liquidity weakness.

Approach and solution

ALCO implements a coordinated response:

1. increases unencumbered HQLA and pre-positions collateral;
2. replaces part of short-term corporate funding with term debt;
3. limits concentrated uninsured deposits;
4. introduces deposit-segment runoff and beta models;
5. reprices new lending using maturity-sensitive FTP;
6. adds floating-rate assets and uses appropriately governed interest-rate swaps to reduce duration exposure; and
7. tests operational access to emergency funding facilities.

After reconfiguration, an internal stress test reports HQLA of 260 million and net 30-day outflows of 250 million, giving an LCR of 104%. The one-year negative repricing gap falls from 180 million to 80 million, reducing the estimated adverse NII effect of a 2% increase from 3.6 million to 1.6 million.

Learning points

Regulatory compliance before stress did not imply resilience. Liquidity and interest-rate exposures arose from the same strategy: funding long-duration assets with short-term, behaviorally unstable liabilities. The appropriate response combined balance-sheet restructuring, hedging, funding diversification and stronger governance.

10. Governance, Technology and Contemporary Developments

Liquidity and ALM models depend on accurate, timely and granular data. Modern systems integrate core banking records, securities, derivatives, collateral, customer behavior and payment flows. Intraday dashboards can detect unusual withdrawals and collateral movements.

Machine-learning models may improve deposit segmentation and runoff forecasting, but they create model-risk concerns. Historical training data may omit severe crises, customer behavior may change after digital-media events, and opaque models may be difficult to challenge. Human oversight, explainability, validation and conservative stress overlays remain essential.

Contemporary liquidity management increasingly addresses:

• faster digital withdrawals and social-media amplification;
• climate-related collateral and funding shocks;
• cyber incidents that disrupt payment access;
• central clearing and margin requirements;
• foreign-currency liquidity shortages; and
• contagion through common asset holdings.

Best practice is to maintain buffers above minimum requirements, diversify funding, measure liquidity by entity and currency, integrate liquidity with capital planning, validate behavioral assumptions and rehearse the CFP.

Chapter Summary

Liquidity risk arises when a bank cannot meet obligations without unacceptable loss. Funding liquidity concerns access to cash, while market liquidity concerns the ability to transact without substantial price impact. Cash-flow ladders, cumulative gaps, counterbalancing capacity and survival horizons reveal the timing of potential shortages.

Deposit analysis must consider behavioral maturity, concentration, insurance status, runoff and rate sensitivity. The LCR evaluates 30-day resilience, whereas the NSFR addresses structural funding stability. Neither replaces institution-specific stress testing or a credible contingency funding plan.

ALM coordinates funding, liquidity, product pricing and IRRBB. Repricing gaps estimate short-term earnings sensitivity, while EVE measures longer-term economic-value effects. Macaulay duration must be converted to modified duration before use in the standard price-sensitivity approximation. Modified and effective durations are already sensitivity measures and require no further division by one plus yield.

Key Terms

Asset–liability management: Coordinated management of balance-sheet funding, liquidity and interest-rate exposure.

Behavioral maturity: Expected maturity based on customer behavior rather than contractual terms.

Contingency funding plan: Predefined framework for responding to liquidity stress.

Counterbalancing capacity: Cash and assets realistically available to cover stressed outflows.

Cumulative liquidity gap: Sum of net cash-flow gaps through a specified horizon.

Deposit beta: Proportion of a market-rate change passed through to deposit rates.

Duration gap: Difference between asset duration and liability duration adjusted for their relative values.

Economic value of equity: Present value of assets minus present value of liabilities.

Effective duration: Price sensitivity estimated by revaluing an instrument under upward and downward yield shocks.

Encumbered asset: Asset pledged or otherwise unavailable for unrestricted funding use.

Funding liquidity risk: Risk of being unable to meet obligations when due without unacceptable loss.

High-quality liquid assets: Eligible assets expected to remain liquid during stress.

Liquidity Coverage Ratio: Ratio of eligible HQLA to 30-day stressed net cash outflows.

Maturity mismatch: Difference between the timing of asset inflows and liability outflows.

Modified duration: First-order percentage price sensitivity to a yield change.

Net Stable Funding Ratio: Ratio of available stable funding to required stable funding.

Repricing gap: Rate-sensitive assets minus rate-sensitive liabilities for a specified horizon.

Survival horizon: Period for which available liquidity covers stressed cumulative cash needs.

Review Questions

Short Answer Questions

1. Distinguish funding liquidity risk from market liquidity risk.
2. What does a negative cumulative cash-flow gap indicate?
3. Why can the behavioral maturity of a deposit differ from its contractual maturity?
4. Define deposit beta.
5. What is the purpose of a contingency funding plan?

Long Answer Questions

1. Explain how funding and market liquidity can interact during a banking crisis.
2. Discuss the construction, interpretation and limitations of a maturity ladder.
3. Compare the LCR and NSFR in terms of horizon, purpose and limitations.
4. Explain how ALCO, funds-transfer pricing and limits support effective ALM.
5. Compare repricing-gap analysis with the EVE approach to IRRBB.

Analytical / Application Questions

1. A bank has HQLA of 180 million, 30-day outflows of 260 million and eligible inflows of 90 million. Calculate its LCR under the 75% inflow cap and interpret the result.
2. A bank has rate-sensitive assets of 650 million and rate-sensitive liabilities of 780 million. Estimate the annual NII effect of a 1.25 percentage-point rate increase, stating the assumptions.
3. A bank has assets of 1,500 million with modified duration 4.0 and liabilities of 1,350 million with modified duration 1.8. Estimate the change in EVE following a parallel 150-basis-point increase and discuss the limitations of the estimate.

Multiple Choice Questions with Answers

Question 1

Which statement best describes funding liquidity risk?

A. The risk that an asset’s credit rating will decline

B. The risk that obligations cannot be met when due without unacceptable loss

C. The risk that market interest rates will remain unchanged

D. The risk that accounting income exceeds taxable income

Correct Answer: B. The risk that obligations cannot be met when due without unacceptable loss

Question 2

A bank records inflows of 70 million and outflows of 95 million in a time bucket. What is its net liquidity gap using G_t=I_t-O_t?

A. +165 million

B. +25 million

C. −25 million

D. −165 million

Correct Answer: C. −25 million

Question 3

A bank has HQLA of 120 million, outflows of 180 million and eligible inflows of 60 million. The inflow cap does not bind. What is its LCR?

A. 66.7%

B. 100.0%

C. 150.0%

D. 200.0%

Correct Answer: B. 100.0%

Question 4

What does an NSFR of 105% indicate?

A. Available stable funding is 5% less than required stable funding

B. HQLA exceeds 30-day outflows by 5%

C. Available stable funding is 1.05 times required stable funding

D. Loans exceed deposits by exactly 5%

Correct Answer: C. Available stable funding is 1.05 times required stable funding

Question 5

A market rate rises by 2 percentage points and a bank’s deposit rate rises by 0.5 percentage points. What is the deposit beta?

A. 0.25

B. 0.50

C. 2.50

D. 4.00

Correct Answer: A. 0.25

Question 6

A bank has rate-sensitive assets of 500 million and rate-sensitive liabilities of 620 million. Under the simple repricing-gap model, what is the approximate annual NII effect of a 1% rate increase?

A. +6.20 million

B. +1.20 million

C. −1.20 million

D. −6.20 million

Correct Answer: C. −1.20 million

Question 7

Assets are 1,000 million with modified duration 3.8, liabilities are 900 million with modified duration 1.6, and yields rise by 1%. What is the approximate change in EVE?

A. −52.4 million

B. −38.0 million

C. −23.6 million

D. +23.6 million

Correct Answer: C. −23.6 million

Question 8

Which formula correctly defines effective duration for equal upward and downward yield shocks?

A. (P_- - P_+) ÷ (2P₀Δy)

B. (P_+ - P_-) ÷ P₀

C. D_Mac × (1+y)

D. P₀ ÷ (P_- + P_+)

Correct Answer: A. (P_- - P_+) ÷ (2P₀Δy)

Question 9

Which indicator most directly measures the cost of immediately buying or selling a traded asset?

A. Deposit beta

B. Bid–ask spread

C. Loan-to-deposit ratio

D. Repricing gap

Correct Answer: B. Bid–ask spread

Question 10

Why should a contingency funding plan distinguish funding availability from usability?

A. Available funding may be operationally delayed, legally restricted or unavailable during stress

B. All available funding must be recognized as accounting equity

C. Usable funding is always cheaper than deposits

D. Regulatory ratios prohibit secured borrowing

Correct Answer: A. Available funding may be operationally delayed, legally restricted or unavailable during stress