Asset–Liability, Treasury and Market Risk Management

Introduction and context

Financial institutions transform maturities, liquidity, currencies and risks. A bank may fund long-term, fixed-rate loans with short-term or floating-rate deposits; an insurer may receive long-dated premiums while investing in bonds; an investment firm may hold positions whose values change with rates, exchange rates, credit spreads and market volatility. This transformation creates earnings opportunities, but also exposes the institution to interest-rate risk, liquidity risk, currency risk and market risk.

Asset–liability management (ALM) is the governance and analytical process used to manage these exposures in a manner consistent with solvency, liquidity, profitability, regulation and risk appetite. Treasury is the operating centre of this process. It manages funding, cash, market positions, liquidity buffers, collateral and hedging, while the asset–liability committee (ALCO) establishes policies and limits.

The central managerial challenge is not to eliminate risk. Complete elimination may be costly or may remove economically valuable intermediation. The objective is to identify the institution’s risk capacity, measure exposures under realistic assumptions, select appropriate mitigants and ensure that residual risks are deliberate, priced and governed.

Learning objectives

By the end of this chapter, the reader should be able to:

1. Explain the relationship between ALM, treasury and market-risk governance.
2. Calculate and interpret maturity gaps, repricing gaps, duration gaps and liquidity metrics.
3. Explain funds-transfer pricing and its role in product profitability.
4. Evaluate hedging choices using swaps, futures, options and foreign-exchange contracts.
5. Apply value-at-risk, stress testing and scenario analysis while recognising their limitations.
6. Design an integrated response to interest-rate, liquidity, currency and market exposures.
7. Assess governance, regulatory, ethical and technological requirements for effective risk management.

1. The ALM architecture

1.1 Balance-sheet transformation

A simplified bank balance sheet contains assets such as loans, securities and cash, and liabilities such as deposits, wholesale borrowings and issued debt. Equity absorbs unexpected losses. Risk arises because the cash-flow, repricing, currency and behavioural characteristics of assets differ from those of liabilities.

Four distinctions are essential:

• Maturity risk: contractual cash flows occur at different dates.
• Repricing risk: assets and liabilities reset their interest rates at different times or by different amounts.
• Basis risk: instruments use different reference rates, such as a loan linked to one benchmark and funding linked to another.
• Optionality risk: customers can prepay loans, withdraw deposits or exercise embedded options, changing expected cash flows.

ALM therefore uses both contractual data and behavioural assumptions. A demand deposit may have no contractual maturity but can provide stable funding. A mortgage may have a long legal maturity but a shorter effective life because of prepayments.

1.2 Governance structure

The board approves risk appetite and liquidity strategy. ALCO translates that appetite into limits for gaps, duration, liquidity, funding concentration, foreign exchange and stress losses. Treasury executes funding and hedging within those limits. Independent risk management measures exposures and challenges assumptions. Internal audit evaluates the control framework. Finance validates accounting treatment and performance measurement.

A sound framework separates:

1. Risk taking—business units originate or trade positions.
2. Risk control—treasury and risk functions aggregate and hedge exposures.
3. Independent oversight—risk committees, compliance and internal audit test decisions.

This separation reduces conflicts of interest, such as a desk delaying loss recognition to preserve reported performance.

2. Measuring interest-rate exposure

2.1 Repricing and maturity gaps

A repricing gap compares rate-sensitive assets and liabilities within a time bucket. Let RSAᵢ be rate-sensitive assets repricing in bucket i and RSLᵢ be rate-sensitive liabilities repricing in that bucket:

A positive gap means more assets than liabilities reprice. If rates rise and the repricing is approximately simultaneous, net interest income tends to increase. A negative gap tends to benefit from falling rates. The approximation for the change in net interest income is:

where ΔNIIᵢ is the change in interest income less interest expense over the measurement horizon, Gᵢ is expressed in currency units, and Δrᵢ is the rate change expressed as a decimal. The approximation assumes parallel, immediate repricing, unchanged balances, no basis risk and no customer behaviour changes. These assumptions are often unrealistic.

A maturity gap instead compares contractual principal cash flows by maturity. It is useful for refinancing analysis but does not adequately measure instruments that reprice before principal repayment. A five-year floating-rate loan may have a five-year contractual maturity but a short interest-rate repricing interval.

Illustrative calculation. Suppose a bank has RSA of $420 million and RSL of $500 million in the three-to-six-month bucket. The gap is −$80 million. If relevant rates rise by 1 percentage point, then:

The estimated annualised effect is a $0.8 million decline, before tax and behavioural effects. The result is not a forecast; it is a sensitivity indicator.

2.2 Duration and economic value

Duration measures the sensitivity of a bond’s price to yield. Macaulay duration is the present-value-weighted average time to cash flow:

where t is the time in years, CFₜ is the cash flow at time t, PV(CFₜ) is its present value, and P is the instrument’s price. Modified duration is:

where y is the annualised yield and m is the number of compounding periods per year. For a small yield change, the price effect is:

This linear approximation assumes a small, parallel yield change and ignores convexity. For larger changes, convexity should be included:

where C is convexity, measured consistently with the duration convention.

For a balance sheet, the duration gap focuses on economic value of equity (EVE). A simple measure is:

where D_A and D_L are the duration of assets and liabilities, and A and L are their market values. If DG is positive, a rate increase generally reduces asset value more, relative to the liability structure, and may reduce EVE. The exact effect is better expressed as:

This assumes equal yield changes and market-valued instruments. Deposits with administered or behavioural rates require effective duration estimates rather than contractual maturity alone.

Worked example. Assets have market value $100 million and duration 4.2 years. Liabilities have market value $90 million and duration 1.5 years. For a 1% parallel increase:

The estimated EVE decline is $2.85 million. Convexity, credit spreads, deposits, mortgage prepayments and hedging would alter the result.

2.3 Yield-curve, basis and optionality risk

A parallel-shift analysis is insufficient when short-term and long-term rates move differently. A bank may be asset-sensitive at short maturities but liability-sensitive at longer maturities. Key-rate duration measures sensitivity to specific maturity points and supports non-parallel scenario analysis.

Basis risk occurs when cash flows reference different rates. A three-month loan tied to one benchmark cannot be perfectly hedged by borrowing tied to another benchmark. The spread between benchmarks may widen during stress.

Optionality is particularly important in mortgages, callable bonds and deposits. Falling rates may increase loan prepayments just when the institution expects to retain high-yielding assets. Deposit betas—the proportion of a market-rate change passed to depositors—may differ between normal and stressed conditions. ALM models should therefore use documented behavioural assumptions, back-testing and conservative sensitivity ranges.

3. Liquidity and funding risk

Liquidity risk is the inability to meet obligations when due without unacceptable loss. It has both funding liquidity and market liquidity dimensions. Funding liquidity concerns cash availability; market liquidity concerns the ability to sell or finance an asset without materially moving its price.

3.1 Core liquidity metrics

The liquidity coverage ratio (LCR) is conceptually:

HQLA means high-quality liquid assets after applicable regulatory adjustments. Net cash outflows incorporate expected outflows less permitted inflows under the relevant regulatory framework. The ratio is not simply total cash divided by deposits; its definitions are jurisdiction-specific.

The net stable funding ratio (NSFR) compares stable funding available with stable funding required over a longer horizon:

Stable funding weights depend on liability stability and asset liquidity characteristics. Internal metrics should supplement regulatory ratios, including survival horizon, unencumbered collateral, concentration of funding, intraday liquidity and stressed cash-flow gaps.

A cumulative liquidity gap over horizon T can be represented as:

A negative value signals a funding need, but the conclusion depends on access to secured funding, liquid assets and contingency sources.

3.2 Liquidity stress testing

Stress tests should combine institution-specific and market-wide events: deposit run-off, loss of wholesale funding, collateral haircuts, margin calls, rating downgrade, market closure and reduced asset-sale prices. Reverse stress testing asks which combination of events would exhaust the liquidity buffer or breach a survival horizon.

A contingency funding plan should identify early-warning indicators, escalation authorities, eligible collateral, central-bank access where applicable, communication protocols and recovery actions. Liquidity cannot be managed effectively by ratios alone; timing, operational capacity and legal transferability of cash matter.

4. Treasury and funds-transfer pricing

Treasury manages the institution’s central funding, liquidity buffer, interest-rate position, currency position and market access. A treasury function may centralise risk or allocate it to business units, but responsibilities must be explicit.

Funds-transfer pricing (FTP) assigns an internal price to the funding and liquidity consumed or provided by a product. A matched-maturity approach prices a loan against a transfer rate corresponding to its expected funding tenor, rather than using a single corporate average funding cost.

A simplified product contribution is:

For a deposit, treasury may pay the business unit an FTP credit for stable funding, adjusted for liquidity value, optionality and behavioural life. FTP should include:

• reference funding curve;
• liquidity premium and contingency value;
• optionality and behavioural adjustments;
• currency and basis costs;
• regulatory balance-sheet costs where relevant;
• transparent allocation rules.

FTP is a management-accounting system, not necessarily an external accounting measure. Poor FTP can reward long-term loans funded by unstable short-term resources or overstate the profitability of volatile deposits. Independent validation is needed because FTP assumptions influence pricing, sales incentives and reported performance.

5. Hedging and derivatives

5.1 Instruments and uses

A forward fixes a future exchange rate or price. A futures contract is standardised and exchange-traded, with daily margining. A swap exchanges cash-flow types, such as fixed interest for floating interest or one currency for another. An option gives the buyer a right but not an obligation, in exchange for a premium.

Interest-rate swaps can convert fixed-rate borrowing into floating-rate exposure or the reverse. A receive-fixed, pay-floating swap generally benefits when rates fall, while a pay-fixed, receive-floating swap generally benefits when rates rise. The economic result also depends on the underlying position and the institution’s accounting treatment.

Currency swaps and forwards hedge foreign-currency assets, liabilities and expected cash flows. Options can preserve favourable upside while limiting adverse movements, but premiums and valuation complexity may be substantial.

5.2 Hedge design

A hedge should begin with a documented risk objective: reduce earnings volatility, protect EVE, lock a funding cost, hedge a foreign-currency commitment or control trading risk. The institution should then identify the exposure, hedge ratio, horizon, eligible instruments, basis risks, counterparty risks and termination conditions.

For a simple futures hedge, the number of contracts can be approximated by:

where N is the hedge quantity, D_P and D_F are price sensitivities or durations of the portfolio and futures instrument, and V_P and V_F are their respective market-value scales. The formula assumes a stable relationship and should be adjusted for contract specifications and hedge direction. It is not a universal hedge ratio.

Derivative risks include counterparty default, collateral and margin liquidity, model risk, basis risk, legal enforceability, settlement risk and misuse of leverage. Netting agreements and collateral reduce, but do not eliminate, counterparty exposure.

6. Market risk measurement

6.1 Value-at-risk

Value-at-risk (VaR) estimates a loss threshold over a specified holding period and confidence level. A statement such as “one-day 99% VaR of $5 million” means that, under the model and assumptions, losses are expected to exceed $5 million on approximately 1% of comparable days. It does not mean the maximum possible loss.

For a single normally distributed position with value V, volatility σ per day and confidence quantile zα, a simplified VaR is:

For a 99% confidence level under a normal distribution, zα is approximately 2.326. The formula assumes linear exposure, normal returns, stable volatility and no liquidity adjustment. Portfolio VaR requires correlations and a correctly specified covariance structure. Historical simulation avoids a normality assumption but depends on the historical window. Monte Carlo simulation can represent nonlinear derivatives but is computationally and model intensive.

VaR is useful for common measurement, limit setting and capital allocation. Its weaknesses include tail blindness beyond the selected percentile, procyclicality, dependence on data and model assumptions, and weak treatment of liquidity and regime changes. Expected shortfall, the average loss conditional on exceeding VaR, provides more information about the tail, although it also depends on model quality.

6.2 Stress and scenario analysis

Stress testing asks what happens under specified shocks rather than relying only on probability estimates. Examples include a 300-basis-point yield-curve steepening, a 15% currency depreciation, a widening of credit spreads, a 20% decline in equity prices, a deposit outflow and simultaneous collateral calls.

Scenario analysis should be coherent: rates, exchange rates, volatility, funding spreads and customer behaviour should move consistently with the scenario. Sensitivity analysis changes one variable at a time; scenario analysis changes several linked variables. Reverse stress testing identifies conditions that make the business model fail.

7. Integrated management process

An effective cycle is:

1. Identify: map contractual and behavioural cash flows, currencies, benchmarks, collateral and embedded options.
2. Measure: calculate gaps, duration, EVE, NII sensitivity, liquidity ratios, VaR and stress losses.
3. Set limits: establish board-approved thresholds, escalation triggers and risk appetite statements.
4. Price: reflect funding, liquidity, capital, optionality and hedging costs through FTP and customer pricing.
5. Mitigate: rebalance assets and liabilities, raise funding, hold liquid assets or execute hedges.
6. Monitor: report daily or intraday exposures, limit usage, model exceptions and emerging concentrations.
7. Validate: back-test models, challenge assumptions, conduct independent review and update scenarios.

Technology supports this process through integrated position databases, intraday cash forecasting, automated collateral management, independent pricing, model inventories and dashboards. Artificial intelligence may improve anomaly detection and behavioural forecasting, but opaque models create explainability, data-quality, bias and model-risk concerns. Human oversight, challenger models and documented override authority remain essential.

8. Illustrative integrated case: Meridian Commercial Bank

Background

Meridian Commercial Bank has $1 billion of assets. Its loan book is 60% fixed-rate with an average duration of 4.5 years. It is funded by $700 million of retail deposits with an estimated effective duration of 1.2 years and $200 million of six-month wholesale funding. The remaining balance is equity and liquid assets. The bank also holds €80 million of euro-denominated securities funded partly in dollars.

Problem

Management expects rates to rise, wholesale markets to become less reliable and the domestic currency to depreciate. A simple repricing report shows a negative short-term gap, while the duration analysis indicates a substantial EVE decline under a parallel rate increase. The bank also faces foreign-exchange and liquidity risk.

Analysis

The ALCO first separates risks. The short-term negative gap means interest expense may reset faster than loan income. The duration mismatch indicates that the market value of assets is more rate-sensitive than liabilities. The euro securities create a foreign-currency asset exposure. Wholesale funding creates refinancing and rollover risk.

The bank runs a 200-basis-point rate shock, a three-month wholesale-market closure, a 10% deposit outflow, a 5% haircut increase on collateral and a 10% domestic-currency depreciation. It estimates both NII and EVE effects, including deposit beta and prepayment assumptions. VaR is reported for ordinary market conditions, but the stress scenario is used for solvency and liquidity decisions.

Approach and solution

Meridian takes five actions:

1. It enters pay-fixed, receive-floating interest-rate swaps against a defined portion of fixed-rate assets, reducing duration exposure while retaining a measured amount of rate risk.
2. It extends part of wholesale funding and increases committed backup lines, while avoiding excessive concentration in one market.
3. It raises the stock of unencumbered liquid assets and tests whether collateral can be mobilised operationally.
4. It hedges the euro securities with forwards, matching notional amount, maturity and expected cash flows while monitoring rollover and basis risk.
5. It revises FTP so that long-duration loans bear the cost of stable funding, liquidity and hedging; stable retail deposits receive an appropriate funding credit.

Limits are established for swap counterparty exposure, hedge ineffectiveness, currency open position, cumulative cash-flow gaps and stress losses. Treasury reports daily, while ALCO reviews assumptions monthly or when market conditions change.

Outcome and learning points

The hedges reduce EVE sensitivity and currency volatility, but they create margin, collateral and counterparty obligations. Longer funding improves liquidity resilience but increases cost. The case demonstrates that no single metric is sufficient: gap analysis identifies repricing timing, duration measures value sensitivity, liquidity metrics assess survival, FTP affects incentives, and stress testing exposes interactions among risks.

9. Limitations, ethics and regulation

Models can create false confidence. Contractual maturity may not represent customer behaviour; correlations can break down; historical data may omit crises; derivative valuations may depend on unobservable inputs; and liquidity can disappear precisely when models assume it is available. Model risk should therefore be governed through validation, benchmarking, sensitivity analysis, performance monitoring and clear ownership.

Ethically, risk managers must avoid hiding losses through optimistic assumptions, manipulating limits, selecting favourable scenarios or selling products whose risks customers cannot understand. Senior managers should ensure that remuneration does not reward short-term income while transferring tail risk to depositors, taxpayers or future shareholders.

Regulatory expectations commonly emphasise capital adequacy, liquidity buffers, interest-rate risk in the banking book, stress testing, governance, recovery planning, market-conduct controls, derivative margining and accurate disclosures. Exact requirements differ by jurisdiction and should be checked against current supervisory rules. Compliance, however, is a floor rather than a complete risk strategy.

Chapter Summary

ALM manages the economic consequences of transforming maturities, rates, currencies and liquidity. Repricing gaps provide a short-horizon view of possible net-interest-income sensitivity, while maturity gaps focus on contractual cash-flow timing. Duration and EVE analysis assess the market-value effect of interest-rate changes, but require assumptions about yield curves, deposits and optionality. Liquidity management combines LCR, NSFR, cash-flow gaps, survival horizons, collateral and contingency funding plans. Treasury executes funding, liquidity and hedging decisions, while FTP allocates the economic cost and value of funding and liquidity to products. Derivatives can reduce exposures but introduce basis, counterparty, collateral and model risks. VaR supports ordinary-market measurement, whereas stress, scenario and reverse-stress analysis are essential for tail events. Effective practice integrates measurement, pricing, limits, hedging, governance, technology and ethical judgment.

Key Terms

Asset–liability management: Coordinated management of balance-sheet risks, returns and funding.

ALCO: Committee responsible for balance-sheet risk strategy and limits.

Repricing gap: Rate-sensitive assets minus rate-sensitive liabilities in a time bucket.

Maturity gap: Difference between contractual asset and liability cash flows by maturity.

Duration: Present-value-weighted sensitivity of a security’s value to yield changes.

Economic value of equity: Market value of assets less market value of liabilities.

Basis risk: Risk that related reference rates or prices do not move together.

Optionality risk: Risk that embedded customer or contract options alter cash flows.

Liquidity risk: Inability to meet obligations without unacceptable loss.

HQLA: High-quality liquid assets eligible for liquidity-buffer purposes.

LCR: Liquidity coverage ratio for a specified short-term stress horizon.

NSFR: Net stable funding ratio for longer-term structural funding.

Funds-transfer pricing: Internal allocation of funding, liquidity and related costs.

Hedge: Position intended to offset an identified risk exposure.

VaR: Estimated loss threshold at a stated confidence and holding period.

Expected shortfall: Average loss conditional on exceeding a specified VaR threshold.

Stress testing: Analysis of effects under severe specified conditions.

Reverse stress testing: Identification of scenarios that could cause business failure or limit breach.

Model risk: Risk of adverse decisions arising from incorrect or misused models.

Review Questions

Short Answer Questions

1. Distinguish a repricing gap from a maturity gap.
2. Why can a positive repricing gap benefit from rising interest rates?
3. What does the LCR measure conceptually?
4. Identify two limitations of VaR.
5. Why should FTP include liquidity and optionality costs?

Long Answer Questions

1. Explain how duration analysis complements repricing-gap analysis in managing interest-rate risk.
2. Discuss the design of a comprehensive liquidity stress-testing programme.
3. Explain the economic purposes and principal risks of interest-rate swaps and currency forwards.
4. Evaluate the role of treasury, ALCO, risk management and internal audit in governance.
5. Compare historical, parametric and Monte Carlo VaR approaches and explain when stress testing remains necessary.

Analytical / Application Questions

1. A bank has $300 million of rate-sensitive assets and $380 million of rate-sensitive liabilities in a bucket. Estimate the NII effect of a 1.5% rate increase and interpret the result.
2. Assets of $120 million have duration 5.0 and liabilities of $105 million have duration 2.0. Estimate the first-order EVE change for a 1% parallel rate increase, ignoring convexity.
3. Design a response for a bank facing a 15% deposit outflow, a 250-basis-point rate increase and a 12% depreciation of its domestic currency. Identify measurements, hedges, funding actions and governance controls.

Multiple Choice Questions with Answers

Question 1

A bank has rate-sensitive assets of $420 million and rate-sensitive liabilities of $500 million. If rates rise by 1%, the approximate change in annual net interest income is:

A. +$0.8 million

B. −$0.8 million

C. +$8.0 million

D. −$8.0 million

Correct Answer: B. −$0.8 million

Question 2

Which measure primarily captures the market-value sensitivity of a bond to a small yield change?

A. Modified duration

B. LCR

C. NSFR

D. Funding concentration ratio

Correct Answer: A. Modified duration

Question 3

An institution with more rate-sensitive liabilities than assets in a repricing bucket generally tends to experience what from a rate increase, all else equal?

A. Higher NII

B. Lower NII

C. No change in NII

D. Higher liquidity automatically

Correct Answer: B. Lower NII

Question 4

The conceptual formula for LCR is:

A. Net cash outflows ÷ HQLA

B. HQLA ÷ net cash outflows

C. Capital ÷ total assets

D. Stable funding ÷ required capital

Correct Answer: B. HQLA ÷ net cash outflows

Question 5

Which derivative gives the buyer a right, but not an obligation, to transact?

A. Forward

B. Future

C. Swap

D. Option

Correct Answer: D. Option

Question 6

A one-day 99% VaR of $5 million means that:

A. The maximum possible loss is $5 million

B. The institution will lose exactly $5 million every 100 days

C. Modelled losses are expected to exceed $5 million on about 1% of comparable days

D. The institution has no risk on the other days

Correct Answer: C. Modelled losses are expected to exceed $5 million on about 1% of comparable days

Question 7

Which technique specifically examines the conditions that could cause a business model to fail?

A. Reverse stress testing

B. Matched-maturity FTP

C. Macaulay duration

D. Historical averaging

Correct Answer: A. Reverse stress testing

Question 8

A pay-fixed, receive-floating interest-rate swap generally helps an institution that is exposed to:

A. Rates rising on a floating-rate asset

B. Rates falling on a fixed-rate asset

C. Rates rising on a fixed-rate asset

D. A foreign-currency liability only

Correct Answer: C. Rates rising on a fixed-rate asset

Question 9

If asset duration is 4.2 years and liability duration is 1.5 years, a parallel rate increase will generally have what first-order effect when assets exceed liabilities in value?

A. Increase EVE

B. Reduce EVE

C. Leave EVE unchanged

D. Eliminate liquidity risk

Correct Answer: B. Reduce EVE

Question 10

Which is a principal limitation of a simple parametric VaR model?

A. It cannot state a confidence level

B. It may rely on normality and stable correlations

C. It measures only contractual maturity

D. It automatically includes all liquidity stress

Correct Answer: B. It may rely on normality and stable correlations