Asset Liability Management

Asset-Liability Management (ALM)

When to Use

• Frame ALM objectives: cash-flow matching, surplus protection, return within risk appetite, regulatory capital efficiency
• Analyze interest rate risk: duration, convexity, key rate duration, parallel and non-parallel shocks
• Design liability-driven investment (LDI), immunization, and hedge programs (rates, inflation, FX)
• Quantify duration gap, surplus sensitivity, and surplus-at-risk concepts for ALCO materials
• Support insurer, pension, and bank ALM contexts with institution-appropriate metrics
• Build stress and scenario sets for ALM (rates, spreads, equity, longevity, credit)
• Draft or review ALM policy, risk limits, and ALCO reporting packs (high level)
• Connect ALM to capital, solvency, and regulatory metrics without substituting appointed actuary or risk sign-off
• Explain reinvestment, prepayment, and optionality impacts on asset–liability profiles

When NOT to Use

• Pension plan design, ERISA funding, PBGC, or DB/DC benefit formulas as primary topic → pension-retirement-funds
• Actuarial pricing, reserving, IBNR triangles, mortality table construction, or statutory opinions → actuary
• Assumption governance, assumption packs, and change-control workflows without ALM portfolio lens → assumption-setting
• P&C underwriting, claims, or line-of-business education without ALM balance-sheet focus → property-casualty-insurance
• Life/health product features, distribution, or claims operations without ALM framing → life-health-insurance
• Security selection, issuer research, or equity valuation as primary deliverable → financial-analyst (if installed)
• Actuarial consulting engagement scoping, SOW, or due diligence program management → actuarial-consulting
• Bank intraday liquidity crisis operations, LCR/NSFR runbooks, or treasury payment ops (unless ALM rate-risk context only)
• Trade execution, order management, or portfolio implementation mechanics without ALM risk framing

Related skills

Core Workflows

1. Engagement scoping

Before analysis:

1. Institution type — Insurer (life/P&C), pension fund/trust, bank ALM desk, asset manager LDI mandate 2. Balance sheet — Economic, regulatory, accounting, or funding basis for assets and liabilities 3. Horizon — Short-term liquidity vs long-term solvency; run-off vs going-concern 4. Decision — Hedge design, IPS/ALM policy, ALCO pack, stress test, capital planning input 5. Material risks — Rates, credit/spreads, equity, inflation, longevity, FX, liquidity, basis 6. Governance — ALCO charter, limits, model inventory, independent validation requirements

See `references/alm_scope_and_principles.md`.

2. Interest rate risk and duration

1. Define valuation basis and discount curve(s) for liabilities and assets 2. Compute or interpret effective duration, modified duration, DV01, and convexity 3. Extend to key rate duration and partial durations for non-parallel shocks 4. Quantify duration gap and surplus sensitivity to rate moves 5. Flag embedded options (calls, prepay, guarantees) that break linear duration 6. Coordinate liability cash-flow shapes with actuary or pension-retirement-funds when needed

See `references/interest_rate_risk_and_duration.md`.

3. Liability-driven investing and hedging

1. State LDI objective: minimize surplus volatility, maximize hedge ratio, or cash-flow match 2. Map liability cash flows (timing, indexation, options) to asset segments 3. Select hedge instruments: government bonds, swaps, futures, options, inflation-linked 4. Design immunization or contingent immunization rules and triggers 5. Address reinvestment risk, curve risk, and basis risk between hedge and liability 6. Separate strategic asset allocation from overlay and dynamic de-risking glide paths

See `references/liability_driven_investing.md`.

4. Insurance and pension ALM

1. Identify regime-specific metrics (e.g., surplus, PVFP, economic capital, funded ratio) 2. For insurers: relate ALM to guarantees, asset adequacy, and market risk capital (overview) 3. For pensions: link duration, glide paths, and de-risking to funding and accounting bases 4. For banks: distinguish ALM (IRRBB, EVE/NII) from liquidity risk management 5. Coordinate longevity, lapse, and morbidity with actuarial owners—not duplicate liability models

See `references/insurance_and_pension_alm.md`.

5. Stress testing and governance

1. Define scenario set: historical, hypothetical, regulatory, and reverse stress 2. Shock rates, spreads, equity, credit, inflation, and longevity consistently 3. Report surplus, capital, and limit breaches with clear attribution 4. Align with ALM policy limits, risk appetite, and escalation paths 5. Document model risk, data lineage, and ALCO decision log

See `references/stress_scenarios_and_governance.md`.

6. ALM reporting and metrics

1. Build ALCO dashboard: surplus, duration gap, hedge ratio, key sensitivities 2. Include bridges (market moves, assumption changes, flows, rebalancing) 3. Summarize forward-looking metrics: surplus-at-risk, earnings-at-risk (institution-specific) 4. Tie to capital and regulatory ratios at overview—escalate filings to qualified roles 5. State limitations and basis in every exhibit footnote

See `references/alm_reporting_and_metrics.md`.

Key metrics (ALM)

Always state measurement basis, curve, and rebalancing assumptions.

Data requests (starter checklist)

When the user has not supplied data, ask for:

1. Valuation date and reporting bases (economic, regulatory, accounting, funding) 2. Liability cash-flow projection or summary profile (duration, key rates, inflation linkage) 3. Asset holdings with classification (government, credit, alternatives, derivatives) 4. Existing hedge book (notionals, maturities, counterparties, collateral) 5. ALM policy and risk limits; prior ALCO materials 6. Prior stress results and capital model outputs (overview)

Deliverable standards

Always state uncertainty and limitations. Do not present outputs as investment advice, actuarial opinion, regulatory filing, or legal guidance without qualified human review.

When to load references

• Scope and principles → references/alm_scope_and_principles.md
• Interest rate risk and duration → references/interest_rate_risk_and_duration.md
• LDI, immunization, hedging → references/liability_driven_investing.md
• Insurance and pension ALM → references/insurance_and_pension_alm.md
• Stress, scenarios, governance → references/stress_scenarios_and_governance.md
• Reporting and metrics → references/alm_reporting_and_metrics.md

ALM reporting and metrics

Table of contents

1. Reporting audiences 2. Core ALM dashboard 3. Sensitivity and exposure metrics 4. Tail and horizon metrics 5. Bridges and attribution 6. Benchmarking and IPS alignment 7. Capital and regulatory exhibits 8. Exhibit quality checklist

Reporting audiences

Tailor depth; avoid mixing bases without reconciliation tables.

Core ALM dashboard

Recommended panels (institution-dependent):

1. Surplus or funded ratio — level and 12-month trend 2. Duration gap — total and key-tenor breakdown 3. Hedge ratio — by risk factor (rates, inflation, FX) 4. Limit utilization — traffic-light vs policy 5. Largest sensitivities — top three drivers of \(\Delta S\) 6. Upcoming cash flows — benefit/claim payments vs asset maturities (12–24m) 7. Derivative exposure — notional, DV01, collateral

Include as-of date, currency, and basis in header/footer.

Sensitivity and exposure metrics

Present gross and hedged columns where overlays exist.

Tail and horizon metrics

Always disclose:

• Confidence level (95%, 99%)
• Horizon (1y, 3y)
• Rebalancing assumptions
• Correlations source

Bridges and attribution

Surplus bridge (example components)

\[ S_1 - S_0 = \underbrace{\text{market}}_{\text{rates, spreads, equity}} + \underbrace{\text{flows}}_{\text{contributions, benefits}} + \underbrace{\text{assumptions}}_{\text{actuarial}} + \underbrace{\text{other}}_{\text{FX, fees}} \]

Funded ratio bridge (pensions)

• Asset return
• Liability interest cost / discount rate change
• Contributions and benefit payments
• Demographic experience

Require consistent basis across periods; flag one-offs.

Benchmarking and IPS alignment

ALM reporting should connect to Investment Policy Statement (IPS):

• Strategic asset allocation ranges vs actual
• LDI allocation vs policy target
• Tracking error to liability benchmark (if used)
• Risk budget consumption (equity, credit, illiquids)

Distinguish policy benchmark from peer comparison—peers optional for ALCO, not core ALM risk.

Capital and regulatory exhibits

High-level mapping only—specialists own filings.

Footnote: not substitute for appointed actuary, auditor, or regulatory submission.

Exhibit quality checklist

Before distributing ALCO materials:

• [ ] Basis and currency on every page
• [ ] Curve IDs and shock definitions match risk system
• [ ] Net exposures shown, not assets alone
• [ ] Hedged vs unhedged clearly separated
• [ ] Limits with utilization % and trend
• [ ] Scenarios dated and versioned
• [ ] Narrative explains breaches and proposed actions
• [ ] Disclaimers on non-advice and model limitations
• [ ] Data cut timestamp and reconciliation to finance records

Coordinate liability exhibits with actuary or pension-retirement-funds owners before external distribution.

ALM scope and principles

Table of contents

1. What ALM is 2. Objectives and tradeoffs 3. Stakeholders and governance 4. Measurement bases 5. Risk taxonomy for ALM 6. Boundaries with adjacent disciplines 7. Engagement checklist

What ALM is

Asset-liability management (ALM) is the practice of managing the joint behavior of assets and liabilities so that the institution can meet its obligations and strategic goals within risk appetite and capital constraints.

ALM is not synonymous with:

• Portfolio management alone (alpha, security selection) without liability context
• Treasury operations alone (cash positioning, payments) without balance-sheet risk framing
• Actuarial valuation alone without investment and hedge strategy

Core ALM questions:

1. How do interest rates, credit spreads, equity markets, and inflation move surplus (or economic value)? 2. Are cash flows from assets sufficient and appropriately timed for liability outflows? 3. What hedges and investment policies keep risk within limits through stress scenarios?

Objectives and tradeoffs

Institutions often pursue multiple objectives simultaneously. Make tradeoffs explicit.

Risk appetite translates objectives into quantitative limits (duration gap, SaR, hedge ratio floors, concentration limits).

Stakeholders and governance

ALM deliverables should be decision-ready: options, sensitivities, limit usage, and recommended actions—not data dumps.

Measurement bases

ALM analysis must state which basis applies. Mixing bases without reconciliation is a common failure mode.

Document discount curves, credit assumptions, and whether liquidity premiums are embedded in liabilities.

Risk taxonomy for ALM

Interest rate and curve risk

• Level (parallel shifts)
• Slope and curvature (key rate risks)
• Reinvestment risk when asset proceeds must be reinvested at unknown rates
• Prepayment and extension risk on callable or mortgage-related assets

Credit and spread risk

• Migration and default on corporate and structured holdings
• Basis between hedge instruments and liability discount curves

Market risk beyond rates

• Equity beta on surplus when equities are material
• Real estate and alternatives valuation lag and liquidity
• FX when liabilities or hedges are in different currencies

Insurance and pension-specific

• Longevity / mortality (pensions, annuities)
• Lapse, surrender, and utilization (life/health)
• Inflation linkage on indexed benefits or claims

Liquidity and operational

• Ability to meet collateral calls on derivatives
• Fire-sale risk under stress
• Model and data operational risk

Boundaries with adjacent disciplines

Engagement checklist

Before producing ALM analysis, confirm:

• [ ] Institution type and regulatory context identified
• [ ] Valuation date and bases for assets and liabilities
• [ ] Surplus definition agreed (economic vs accounting vs funding)
• [ ] Horizon for risk metrics (instantaneous shock vs 1y VaR/SaR)
• [ ] Hedgeable risks vs retained risks documented
• [ ] ALM policy limits and escalation paths referenced
• [ ] Owners for liability model and investment book data named
• [ ] Outputs labeled not actuarial opinion, legal advice, or trade instruction

Insurance and pension ALM

Table of contents

1. Institution comparison 2. Life and annuity ALM 3. P&C and health ALM (overview) 4. Pension and retirement fund ALM 5. Bank ALM (IRRBB overview) 6. Capital and regulatory touchpoints 7. Coordination with actuarial owners

Institution comparison

Use institution-appropriate vocabulary in ALCO materials.

Life and annuity ALM

Liability features driving ALM

• Guaranteed minimum rates and crediting strategies
• Surrender and lapse options
• Annuitization and GMxB-style guarantees (overview)
• Longevity on payout annuities

Asset strategies

• Liability-sensitive segments: long bonds, swaps, ILBs
• Spread assets: corporate credit with capital and liquidity limits
• Separate accounts vs general account ring-fencing

ALM metrics (overview)

Detailed reserving, pricing, and assumption work → actuary, life-health-insurance.

P&C and health ALM (overview)

P&C liabilities are often shorter and inflation-linked (claims trends) compared to life.

ALM focus areas:

• Loss reserve discounting (where permitted) and investment income matching
• Inflation sensitivity on long-tail lines
• Liquidity for large cat events
• Asset adequacy testing concepts (high level)

Do not replace reserve adequacy analysis with duration metrics alone → actuary, property-casualty-insurance.

Pension and retirement fund ALM

Liability profile

• DB plans: long duration, sensitive to discount rate, mortality, salary inflation
• De-risking path: equity-heavy → LDI → buyout readiness

Funding vs economic ALM

Pension-specific metrics

• Funded ratio (specify numerator/denominator)
• Surplus volatility under rate scenarios
• Contribution risk and policy corridor

Plan design, ERISA, and risk transfer structures → pension-retirement-funds; ALM consumes cash-flow and duration outputs.

Bank ALM (IRRBB overview)

Banks separate:

• Interest rate risk in the banking book (IRRBB) — EVE and NII metrics
• Liquidity risk — LCR/NSFR (not ALM core unless user asks liquidity crisis ops)

Typical metrics

ALM skill covers IRRBB framing for balance-sheet dialogue—not payment operations or crisis liquidity runbooks.

Capital and regulatory touchpoints

High-level links only—qualified roles sign filings.

Coordinate assumption sets with assumption-setting and liability measurement with actuary.

Coordination with actuarial owners

ALM skill interprets and structures investment/hedge responses; it does not replace appointed actuary or actuarial certification.

Interest rate risk and duration

Table of contents

1. Concepts and definitions 2. Duration measures 3. Convexity and optionality 4. Key rate duration 5. Duration gap and surplus sensitivity 6. Curve construction and consistency 7. Common pitfalls 8. Worked patterns (conceptual)

Concepts and definitions

Interest rate risk is the potential change in value or cash flows due to changes in yield curves and related market variables.

For ALM, focus on net exposure: assets minus liabilities (surplus), not asset duration alone.

Clarify whether shocks are instantaneous (valuation impact) or over a horizon with reinvestment (ALM simulation).

Duration measures

Modified duration

Approximate percentage price change for a small parallel yield change:

\[ \Delta P/P \approx -D_{\text{mod}} \cdot \Delta y \]

Used for fixed-rate bonds without embedded options when yields are well-defined.

Effective duration

Duration from numerical revaluation under curve shocks—required when cash flows change with rates (options, structured products, some liabilities).

\[ D_{\text{eff}} = -\frac{P_{-} - P_{+}}{2 P_0 \cdot \Delta y} \]

Macaulay duration

Weighted average time to cash flows; linked to modified duration via yield for simple bonds. Useful for immunization intuition (match Macaulay durations for a single liability payment).

Dollar duration

\(D_{\$} = D \times P\) — useful for hedge sizing in dollar space.

Convexity and optionality

Convexity captures curvature: for large rate moves, linear duration understates gains when rates fall (positive convexity) on plain bonds.

\[ \Delta P/P \approx -D \cdot \Delta y + \frac{1}{2} C \cdot (\Delta y)^2 \]

Embedded options (calls, puts, prepayment, policyholder options) create negative convexity and extension/contraction risk—duration alone is insufficient.

ALM implications:

• Report effective duration and, where material, convexity or full grid revaluation
• Stress non-linear products with scenario grids, not only parallel ±100bp

Key rate duration

Key rate duration (KRD) measures sensitivity to changes at specific curve tenors while holding other key points fixed (method varies by vendor).

Use KRD when:

• Liabilities are long and pension-like (sensitivity at 10y–30y)
• Hedges use butterflies or barbells rather than parallel swaps
• Steepening/fl flattening scenarios dominate risk

Present KRD tables for assets, liabilities, and net with the same curve methodology.

Duration gap and surplus sensitivity

Duration gap (informal):

\[ \text{Gap} \approx D_A \cdot A - D_L \cdot L \]

Definitions vary: some use dollar duration on market values; pensions may use funded weights.

Surplus duration (economic):

\[ D_{\text{surplus}} \approx \frac{D_A \cdot A - D_L \cdot L}{A - L} \]

when surplus \(S = A - L\) is positive and liabilities are measured on a consistent market basis.

Interpretation:

• Positive net duration → surplus falls when rates rise (if liabilities are longer duration than assets)
• Sign depends on relative durations and weights, not asset duration alone

Always pair with convexity and key-rate profile for material books.

Curve construction and consistency

ALM requires one coherent framework:

1. Risk-free or swap curve for discounting (policy choice) 2. Credit spreads for corporate holdings—separate from liability curve unless liability is credit-based 3. Inflation curve for real-linked cash flows (breakeven or RPI/CPI-specific)

Basis risk arises when:

• Liabilities are discounted on corporate bond yields but hedged with government swaps
• OIS vs LIBOR/SOFR legacy conventions differ across books
• Smoothed funding curves vs market hedge curves (pensions)

Document curve IDs, valuation timestamps, and parallel vs bucketed shock conventions.

Common pitfalls

Worked patterns (conceptual)

Parallel shock bridge

1. Start surplus \(S_0\) 2. Apply +100bp parallel to assets and liabilities (full revaluation preferred) 3. Report \(\Delta S\), implied \(D_{\text{net}}\), compare to duration estimate

Key-rate steepener

1. +50bp at 2y, −25bp at 30y (example) 2. Show impact on pension liabilities vs barbell hedge

Immunization check

For a single deterministic liability payment at time \(T\):

• Match Macaulay duration and PV of assets to liability
• Monitor rebalancing as time passes and coupons arrive

Hand off liability cash-flow engineering to actuary or pension-retirement-funds when full projections are required.

Liability-driven investing

Table of contents

1. LDI objectives 2. Cash-flow matching vs duration matching 3. Immunization strategies 4. Hedge instruments 5. Hedge ratio and implementation 6. Basis and reinvestment risk 7. Dynamic de-risking and glide paths 8. Implementation phases

LDI objectives

Liability-driven investing (LDI) aligns the asset portfolio structure with liability risk drivers—primarily interest rates, inflation, and (where relevant) longevity—rather than maximizing benchmark-relative return in isolation.

Typical goals:

State whether the mandate is minimum risk (pension de-risking) or return-seeking within surplus risk budget.

Cash-flow matching vs duration matching

Dedicated cash-flow matching

• Hold assets whose coupons and maturities align with liability payment schedule
• Strong for known fixed cash flows; weak when liabilities are uncertain (options, longevity, inflation)

Duration / factor matching

• Match duration, key rates, or factor exposures rather than each cash flow
• Uses pooled instruments (bonds, swaps) with reinvestment of intermediate coupons

Immunization strategies

Classical immunization (Redington): match PV and duration (sometimes convexity) so surplus is insulated to small parallel rate changes.

Contingent immunization:

1. Start in return-seeking portfolio while surplus cushion exceeds threshold 2. Trigger move to immunized portfolio if surplus falls below floor 3. Requires clear governance and execution playbook

Horizon matching:

• As liability horizon shortens, reduce duration of assets (glide)
• Coordinate with pension de-risking milestones

Hedge instruments

Distinguish economic hedge from accounting hedge designation—coordinate with finance; not legal advice.

Hedge ratio and implementation

Hedge ratio examples:

\[ HR = \frac{\text{Dollar duration of hedges}}{\text{Dollar duration of liability rate risk}} \]

or PV-weighted coverage of key rates.

Implementation considerations:

1. Notional sizing vs DV01 targeting 2. Rebalancing bands (e.g., re-hedge when HR drifts ±5%) 3. Transaction costs and market impact 4. Collateral and liquidity under stress 5. Separate account vs pooled implementation (pensions, insurers)

Document allowed instruments and counterparty limits in ALM policy.

Basis and reinvestment risk

Basis risk: hedge instrument does not move 1:1 with liability discount rate.

Sources:

• Credit spread on liability discount vs government hedge
• OIS/LIBOR/SOFR convention differences
• Smoothed actuarial valuation rates vs market hedges

Reinvestment risk: intermediate coupons must be reinvested at unknown future rates—immunization drifts unless rebalanced.

Mitigations:

• Shorter bucket ladders for near benefits
• Key-rate hedges at long tenors
• Overlay swaps with explicit roll policy

Dynamic de-risking and glide paths

Glide path reduces risk as funded status improves or participants age.

Typical triggers:

Link glide paths to pension risk transfer readiness (pension-retirement-funds) without duplicating legal structuring.

Implementation phases

1. Diagnostic — duration gap, KRD, surplus sensitivities, basis inventory 2. Policy — objectives, limits, permitted instruments, HR targets 3. Pilot — partial hedge (e.g., 50% HR) with monitoring 4. Scale — full LDI; integrate with manager mandates and benchmarks 5. Review — quarterly ALCO; annual policy refresh

Do not specify individual trades without user mandate and compliance context—frame as program design.

Stress scenarios and governance

Table of contents

1. Scenario design principles 2. Scenario types 3. Risk factor shocks 4. Reverse stress and escalation 5. ALM policy and limits 6. ALCO process 7. Model risk and validation 8. Documentation standards

Scenario design principles

Effective ALM stress testing is:

• Coherent — joint shocks respect economic plausibility (e.g., rates down + equity up vs crisis correlation)
• Transparent — assumptions documented and versioned
• Actionable — tied to limits, capital, and management decisions
• Comparable — repeatable quarter-over-quarter with bridges

Define:

1. Valuation approach (full revaluation vs sensitivity approximation) 2. Rebalancing behavior (static vs dynamic hedges) 3. Time horizon (instantaneous vs 1-year path)

Scenario types

Combine level shocks with shape shocks (steepener, flattener, butterfly).

Risk factor shocks

Interest rates

• Parallel ±50/100/200bp
• Key-rate bumps by tenor
• Negative rate floor handling

Credit spreads

• Widening by rating bucket
• Migration overlays for structured books

Equity and alternatives

• Equity -20% / -35% with correlation to rates per scenario narrative
• Private asset valuation lag and haircuts

Inflation

• CPI/RPI paths for indexed liabilities and ILBs
• Wage inflation for active pension participants

Insurance / pension-specific

• Longevity improvement shock (+X years)
• Lapse mass lapse or surrender spike
• Morbidity utilization spike (health)

Report impacts on:

• Surplus or economic value
• Funded ratio
• Earnings (if requested, accounting basis)
• Capital ratios (overview)

Reverse stress and escalation

Reverse stress identifies combinations of factors that drive surplus to policy floor or regulatory minimum.

Steps:

1. Start from limit breach definition (e.g., SaR > appetite) 2. Search or narrative-build scenario combinations 3. Identify mitigations: additional hedges, contribution, benefit changes (pension), capital raise

Escalation:

ALM policy and limits

ALM policy should define:

• Objectives and risk appetite
• Permitted instruments and counterparty criteria
• Quantitative limits (examples below)
• Roles and approval authorities
• Review cadence and exception process

Example limit categories (institution-specific):

Limits must align with measurement basis used in risk systems.

ALCO process

Asset-Liability Committee (ALCO) agenda (template):

1. Market update — rates, spreads, inflation, liquidity 2. Surplus / funded status — bridges since prior meeting 3. Risk metrics — duration gap, KRD, SaR, limit utilization 4. Hedge program — performance, roll, proposed changes 5. Stress results — new scenarios, breaches, mitigations 6. Policy exceptions — status and remediation 7. Decisions and actions — owners and dates

Minutes should capture dissent, assumptions, and follow-ups for audit trail.

Model risk and validation

ALM relies on:

• Liability models (actuarial)
• ALM / risk aggregation engines
• Vendor analytics for derivatives and KRD

Governance elements:

• Model inventory with tiering
• Independent validation and periodic revalidation
• Change control for assumptions and code
• Back-testing of sensitivities vs full revaluation where possible

Escalate model materiality to enterprise model risk management policy.

Documentation standards

Each stress cycle should archive:

1. Scenario definitions (shocks, correlations, dates) 2. Input data cut (positions, curves, FX) 3. Model version IDs 4. Output exhibits and limit breach log 5. Management responses and hedge tickets (reference only)

Label outputs as internal management information—not regulatory filing or actuarial opinion unless reviewed by qualified roles.