Repo in ALM: Liquidity Gap, Interest Rate Gap & Funding Strategy
Table of Contents
Repurchase agreements are a key component of wholesale bank funding. But how do they fit into the broader framework of asset and liability management (ALM)? For treasury and risk professionals, understanding repo’s role in managing liquidity gaps and interest rate gaps is essential. This guide explains how banks use repo to bridge funding deficits, take views on the yield curve, and manage the risks that come with maturity transformation.
Repo in the ALM Framework
Asset and liability management (ALM) encompasses the tools and techniques banks use to minimize market risk and liquidity risk while achieving profit objectives. At its core, ALM manages the mismatch between when cash flows in (from assets) and when it flows out (to liabilities).
Pure cash matching — perfectly aligning asset and liability maturities — would eliminate gap risk entirely. But it would also eliminate the spread income banks earn from maturity transformation: borrowing short-term at lower rates and lending long-term at higher rates.
Repo serves as the primary short-term funding tool that enables active gap management. Rather than match every asset with an identical-maturity liability, ALM managers use repo to strategically choose when to mismatch. A bank might fund a portfolio of 5-year bonds with rolling overnight repo, earning the term spread while accepting the rollover and rate risk that comes with it.
This flexibility makes repo central to ALM strategy. Overnight repo provides immediate liquidity; term repo locks in funding costs; open repo offers flexibility with no fixed maturity. The choice depends on the bank’s view on rates, its liquidity requirements, and its risk appetite. For how dealers actively manage these positions, see repo trading strategies.
The Liquidity Gap
The liquidity gap measures the difference between assets maturing and liabilities maturing at each future date. It answers a simple question: at each point in time, will cash inflows (from maturing assets) exceed cash outflows (to repay maturing liabilities), or vice versa?
In this article, a positive gap means assets maturing exceed liabilities maturing — a cash surplus that must be invested. A negative gap means liabilities maturing exceed assets maturing — a cash shortfall that must be funded in the market. Some texts use the opposite convention, so always verify definitions when comparing sources.
ALM managers organize the gap into maturity buckets: 0-1 month, 1-3 months, 3-6 months, 6-12 months, and so on. This gap profile reveals the timing and magnitude of funding needs across the maturity spectrum.
| Maturity Bucket | Assets | Liabilities | Period Gap | Cumulative Gap |
|---|---|---|---|---|
| 0-1 month | 1,900 | 2,000 | -100 | -100 |
| 1-3 months | 1,860 | 2,230 | -370 | -470 |
| 3-6 months | 890 | 670 | +220 | -250 |
| 6-12 months | 740 | 550 | +190 | -60 |
This bank has a negative gap (cash shortfall) in the short term — more liabilities maturing than assets, requiring funding. By 3-6 months, the gap turns positive (surplus). The ALM desk must secure short-term funding and plan to invest the medium-term surplus.
Cumulative Gap Analysis
The cumulative gap is the running sum of period gaps up to a given maturity. It shows the total funding requirement (or surplus) the bank faces across all periods. If a bank is short $50 million at one week and short another $20 million at two weeks, the cumulative gap at two weeks is $70 million — the total amount that must be funded by that date.
Cumulative gap analysis reveals whether short-term surpluses offset later deficits, or whether the bank faces a persistent funding need that requires structural solutions beyond overnight repo.
Stress Liquidity Gaps
A gap profile that works in normal markets may become unmanageable when funding markets freeze. Stress testing applies haircuts to asset values and run-off assumptions to liabilities — modeling what happens when repo counterparties increase margins or refuse to roll positions entirely.
During the September 2019 repo spike, overnight repo rates surged to nearly 10% as banks scrambled for funding. Institutions that relied heavily on overnight repo faced acute stress — a reminder that gaps must be manageable under both normal AND crisis conditions. For regulatory stress scenarios, see LCR and NSFR liquidity ratios.
The Interest Rate Gap
The interest rate gap measures the sensitivity of net interest income (NII) to changes in interest rates. Unlike the liquidity gap, which covers all assets and liabilities, the interest rate gap focuses on rate-sensitive items — those whose interest rates will reset or reprice within a given period.
A positive interest rate gap means the bank has more assets repricing than liabilities. If rates rise, the bank earns more on assets while liability costs remain temporarily fixed — NII increases. The reverse is true for a negative gap.
The ΔNII formula is a first-order approximation. For a 3-month bucket, multiply by 90/360 to annualize correctly. Actual NII sensitivity also depends on repricing timing within the bucket, rate betas (how quickly administered rates follow market rates), embedded options like prepayments, and the shape of the yield curve — not just its level.
Funding Long vs Funding Short
Banks face a fundamental choice in their funding strategy: fund short or fund long.
Funding Short
- Borrow short-term, lend long-term
- Traditional bank model
- Earns spread from upward-sloping yield curve
- Exposed to rising rates
- Lower funding cost but higher rollover risk
Funding Long
- Borrow long-term, lend short-term
- Less common; often defensive
- Earns spread from inverted yield curve
- Exposed to falling rates
- Higher funding cost but locked-in stability
Most banks fund short because the yield curve is typically upward-sloping. But this creates interest rate risk: if short-term rates rise faster than expected, the spread compresses or inverts.
Silicon Valley Bank held $91 billion in held-to-maturity securities — mostly long-duration Treasuries and agency MBS — funded largely with short-term deposits. When the Fed raised rates aggressively in 2022-2023, SVB’s fixed-rate assets lost value while its funding costs rose. The resulting interest rate gap created $15+ billion in unrealized losses. When depositors withdrew funds, SVB faced a classic ALM failure: assets couldn’t be sold without crystallizing losses, and short-term funding evaporated. The bank failed in March 2023.
Creating a Tail
A tail is the period between when funding matures and when the asset it finances matures. If a bank finances a 6-month loan with rolling 1-month repos, it has a 5-month tail — five months of exposure to uncertain future funding costs.
A bank finances a $100 million 6-month corporate loan at 7% using 1-month repo at 5%. The initial spread is 200 basis points.
If the 1-month repo rate rises to 6% at the next rollover, the spread compresses to 100 basis points. If repo rates exceed 7%, the position becomes unprofitable. The bank has accepted this rate risk by choosing to fund short.
Beyond rate risk, creating a tail also exposes the bank to rollover risk — the possibility that repo counterparties refuse to roll the position at any rate — and liquidity risk if markets seize up.
Collateral and Repo Liquidity Management
For banks using repo as a primary funding tool, collateral management is as important as gap management. Repo funding is only available to the extent that the bank has eligible, unpledged securities to post.
Key collateral considerations:
- Eligibility: Not all securities qualify for repo. Treasuries command the tightest haircuts and may trade special when in high demand; lower-quality collateral may not be accepted at all in stressed markets.
- Haircuts: A 2% haircut means $100 million of Treasuries generates only $98 million of funding. Haircuts can widen rapidly in stress.
- Encumbrance: Securities already pledged as collateral (for derivatives, secured lending, or other purposes) are unavailable for repo.
- Margin calls: If collateral values decline, the bank faces margin calls requiring additional securities or cash — precisely when liquidity is most scarce.
- Substitution rights: Some repo agreements allow substituting collateral; others do not. This affects operational flexibility. For legal details, see GMRA master agreement.
A bank with a $10 billion Treasury portfolio can use repo to generate approximately $9.8 billion of liquidity overnight (assuming a 2% haircut on high-quality Treasuries in normal markets). This funds other activities while the bank retains economic exposure to the bonds.
Note: This example is illustrative. Actual liquidity depends on haircut levels, which vary by collateral type, counterparty, market conditions, and tenor.
Comparison: Gap Limits vs Regulatory Ratios
Banks set limits on gap exposures by maturity bucket to control interest rate and liquidity risk. These limits are typically set by the Asset-Liability Committee (ALCO), which includes senior management from Treasury, Finance, and Risk.
Limit structures include:
- Period limits: Maximum gap for each individual bucket (e.g., “6-month gap cannot exceed $10 million”)
- Cumulative limits: Maximum cumulative gap across buckets (e.g., “cumulative 12-month gap cannot exceed $50 million”)
- NII-at-risk limits: Maximum acceptable change in NII under specified rate scenarios
Gap Limits
- Set by: ALCO (internal)
- Purpose: Risk appetite control
- Flexibility: Bank can adjust based on views
- Scope: All maturities, custom buckets
LCR / NSFR
- Set by: Basel Committee / regulators
- Purpose: Minimum regulatory floor
- Flexibility: Must meet at all times
- Scope: Specific stress scenarios (30-day, 1-year)
Gap limits and regulatory ratios are complementary. A bank can have very tight internal gap limits while still comfortably meeting LCR — or vice versa. For regulatory details, see LCR and NSFR explained.
Common Mistakes
1. Confusing liquidity gap with interest rate gap: The liquidity gap covers ALL assets and liabilities; the interest rate gap covers only rate-sensitive items. They measure different risks and require different hedges.
2. Ignoring stress scenarios: A gap profile optimized for normal markets may become unmanageable when funding dries up. Always stress-test gap positions under adverse scenarios.
3. Assuming overnight repo can always be rolled: Repo markets have historically frozen during crises. Bear Stearns collapsed in March 2008 when repo counterparties refused to roll overnight funding; Lehman Brothers faced similar repo runs in September 2008. The September 2019 rate spike and March 2020 dash-for-cash — when dealer balance-sheet constraints prevented normal market functioning — showed that even Treasury repo can become scarce. For basic repo mechanics, see repurchase agreements explained.
4. Not accounting for collateral constraints: Repo funding requires eligible collateral. If securities are encumbered or haircuts widen, available funding shrinks precisely when it’s needed most.
5. Treating haircut assumptions as stable: Haircuts can double or triple in stressed markets. The funding capacity you model in normal times may not exist in a crisis.
6. Setting gap limits without considering yield curve shape: A positive interest rate gap is profitable when rates rise but harmful when they fall. Gap limits should reflect the bank’s rate view and risk tolerance, not just an arbitrary number.
Limitations of Gap Analysis
Gap analysis is a foundational ALM tool, but it has significant limitations:
- Assumes parallel yield curve shifts: The ΔNII formula assumes all rates move by the same amount. In reality, curve twists and steepening/flattening create basis risk.
- Static snapshot: Gap reports capture a moment in time. New business, prepayments, and withdrawals change the profile daily.
- Ignores optionality: Prepayable loans, callable deposits, and committed credit lines have embedded options that affect cash flow timing in ways gap analysis doesn’t capture.
- Behavioral assumptions required: Demand deposits have no contractual maturity. Modeling their gap contribution requires assumptions about customer behavior that may not hold in stress.
Gap analysis is a starting point, not a complete picture of ALM risk. Banks supplement it with duration analysis, NII simulation, economic value sensitivity, and stress testing to manage interest rate risk comprehensively.
Frequently Asked Questions
Disclaimer
This article is for educational and informational purposes only and does not constitute investment or financial advice. ALM practices, gap limits, and regulatory requirements vary by institution and jurisdiction. The examples and figures cited are illustrative and may not reflect current market conditions or specific institutional practices. Always consult with qualified treasury and risk management professionals for institution-specific guidance.