Quick answer: the Net Stable Funding Ratio (NSFR) asks whether a bank’s assets and off-balance-sheet commitments are supported by funding that is sufficiently stable over roughly a one-year horizon. The ratio is:

NSFR = Available Stable Funding (ASF) / Required Stable Funding (RSF).

Under the Basel standard, the ratio should be at least 100% on an ongoing basis. ASF assigns larger weights to funding sources expected to remain reliable, such as capital and longer-term liabilities. RSF assigns larger weights to assets that are illiquid, long-dated or otherwise require durable funding. The algorithm therefore rewards a bank for matching slow assets with stable liabilities rather than financing long commitments with funding that can disappear quickly.

The LCR asks, “Can we survive the next 30 days?” The NSFR asks, “Is the balance sheet built on funding that can plausibly stay?”

Why this belongs in mathematics

NSFR is weighted balance-sheet mathematics. Every relevant liability, equity item, asset and selected off-balance-sheet exposure is classified, multiplied by a regulatory factor, and aggregated. The difficult part is not division. It is classification: maturity, counterparty type, encumbrance, asset liquidity and contractual structure determine the weights.

The Basel Framework states that the NSFR is intended to reduce overreliance on short-term wholesale funding and promote a more sustainable maturity structure. The US rule similarly describes it as a one-year structural funding measure complementing the shorter-horizon Liquidity Coverage Ratio. See Basel NSF20 and the US NSFR final-rule summary.

1. Available Stable Funding: how much reliable funding does the bank have?

ASF begins with the liability side of the balance sheet. Each funding source is multiplied by an ASF factor reflecting how likely that funding is assumed to remain available over the one-year horizon.

Conceptually:

ASF = Σ liability/equity amount × ASF factor.

Basel factors range from 100% to 0%. Capital and funding with residual maturity of at least one year can receive the highest treatment. Stable retail deposits receive high ASF recognition because they are assumed to be behaviourally more stable than short-term institutional wholesale funding. See Basel NSF30.

2. Funding tenor matters

Suppose two banks each have S$1 billion of funding. Bank A funds itself mainly with five-year debt. Bank B funds itself mainly with overnight unsecured institutional borrowing.

The nominal amount is identical. The structural stability is not. Bank B must continuously refinance. If markets close, tomorrow’s funding becomes today’s problem.

This is why the Basel calibration generally assigns more ASF to longer-term liabilities than shorter-term liabilities. The NSFR is a maturity-transformation constraint disguised as a ratio.

3. Funding type matters too

A three-month retail deposit and three-month financial-institution borrowing do not necessarily receive the same treatment. Retail deposits can be more stable because they are dispersed across many customers and may benefit from deposit-insurance and relationship effects. Institutional wholesale funding can be concentrated and professionally managed, allowing it to move quickly.

In the US NSFR rule, for example, qualifying stable retail deposits can receive a 95% ASF factor while less stable retail categories receive lower factors; short-term funding from financial institutions can receive much less or no ASF depending on maturity and classification. See Federal Reserve section 249.104.

4. Required Stable Funding: which assets need durable support?

RSF looks at the asset and commitment side:

RSF = Σ asset/off-balance-sheet amount × RSF factor.

Assets that are immediately usable or highly liquid require little stable funding. Assets that are long-dated, illiquid or encumbered require more. Basel’s executive summary describes a 0% RSF factor as applying to fully liquid unencumbered assets and a 100% factor as representing assets that need to be entirely financed by stable funding. See BIS NSFR Executive Summary.

5. A simple teaching example

Imagine a simplified bank has weighted ASF of:

S$300m capital and long-term funding;
S$475m recognised stable retail funding;
S$150m recognised less-stable/medium-term funding.

Total ASF = S$925m.

Its weighted RSF from loans, securities, derivatives and commitments totals S$880m.

NSFR = 925 / 880 ≈ 105.1%.

The bank clears a 100% threshold in this teaching example. But a five-percentage-point buffer is not necessarily comfortable. Changes in deposit composition, wholesale maturities, asset encumbrance or loan growth can move both numerator and denominator.

6. Loan growth can consume stable funding even before liquidity stress occurs

Suppose the bank originates another S$100m of long-term illiquid loans funded with S$100m of very short-term wholesale borrowing. The accounting balance sheet still balances. Economically, structural funding weakens.

The new loans add material RSF while the short wholesale liability may contribute little ASF. The NSFR falls. This is exactly the kind of maturity transformation the ratio is designed to make visible.

7. Encumbrance increases the funding burden

An asset pledged for a long period cannot be freely sold or reused for another liquidity need. Basel therefore makes RSF treatment sensitive to encumbrance and remaining encumbrance period.

This connects NSFR to collateral optimisation. The same security can look liquid in a market database and structurally unavailable once it has been pledged.

8. Derivatives make the calculation less intuitive

Derivative positions can create positive replacement values, collateral flows and funding needs. The NSFR therefore contains specific rules for derivative assets, liabilities and variation margin rather than allowing ordinary accounting netting to determine stable-funding treatment automatically.

Basel NSF30 and US section 249.107 separately calculate derivative-related RSF and ASF components. The broader lesson is important: accounting net value and structural funding need are not always the same object.

9. LCR and NSFR are complementary, not substitutes

Question	LCR	NSFR
Horizon	30-day stress	One-year structural funding
Main numerator	HQLA	Available stable funding
Main denominator	Stressed net cash outflows	Required stable funding
Core problem	Short-horizon survival	Sustainable funding structure

A bank can have a strong LCR because it holds a large liquid-asset buffer while still financing too much long-term business with unstable funding. Conversely, a structurally stable bank can still have an immediate 30-day liquidity problem if its stress outflows are extreme.

For the shorter-horizon rule, see How the Liquidity Coverage Ratio Turns a 30-Day Stress Scenario into a Bank Constraint.

10. NSFR creates shadow prices inside the bank

If a new long-term asset consumes scarce RSF capacity, the bank can assign an internal cost to that structural funding use. This cost can enter funds transfer pricing, product pricing and balance-sheet optimisation.

A loan that looks attractive before structural-funding cost may look weaker once the bank recognises the long-term funding resource it consumes. This connects to Funds Transfer Pricing and Balance-Sheet Optimisation.

11. Creative-work lens: a bridge should not depend on temporary scaffolding

Stories about engineering failures often become memorable because the structure looks finished before the support system is truly durable. NSFR asks the banking equivalent. A loan portfolio can look productive and profitable, but what happens if the temporary funding underneath it disappears?

The creative image is only a memory aid. The actual answer comes from liability tenor, depositor behaviour, asset maturity, encumbrance and contractual funding rules.

12. The NSFR algorithmic pipeline

Reconcile liabilities, equity, assets and off-balance-sheet positions.
Classify funding by counterparty, maturity and stability.
Apply ASF factors.
Classify assets by liquidity, maturity and encumbrance.
Apply RSF factors.
Calculate special derivative and SFT treatments.
Add relevant off-balance-sheet RSF requirements.
Sum ASF and RSF.
Calculate NSFR and buffer above the minimum.
Attribute NSFR consumption to businesses and products.
Stress deposit migration, wholesale rollover and asset encumbrance.
Reconcile reported ratios with source systems and disclosures.

13. Failure modes

LCR substitution. A strong 30-day ratio is treated as proof of long-term structural funding strength.
Maturity-bucket errors. Contract dates, behavioural maturities or optional extensions are classified incorrectly.
Retail-stability overconfidence. Deposit behaviour changes but ASF treatment assumptions remain cognitively unchallenged.
Encumbrance blindness. Assets pledged elsewhere remain classified as freely usable.
Derivative netting shortcuts. Accounting netting is incorrectly treated as NSFR funding netting.
Business-growth blindness. New long-term assets consume RSF faster than stable funding grows.
Ratio gaming. Temporary quarter-end transactions improve the reported ratio without improving sustainable structure.
Aggregation blindness. Group-level strength hides a weak currency or legal-entity funding position.

14. Diagnostics and falsifiers

Which liability categories contribute most ASF?
Which asset classes consume most RSF?
What happens if stable deposits migrate into less-stable categories?
What happens if six-month wholesale funding shortens below a regulatory maturity threshold?
How much RSF comes from encumbered assets?
Which products have positive accounting margin but heavy structural-funding consumption?
Does the ratio remain above 100% after one year of planned asset growth?
Can the published NSFR be reconstructed from source balances and classifications?

Suppose someone claims, “The bank is structurally funded because liabilities equal assets.” A falsifier is a balance sheet in which long-dated illiquid assets are financed mainly by short-term funding that must be continuously rolled. Accounting equality is compulsory; funding stability is not.

15. Verification and update triggers

reconcile regulatory classifications to general-ledger and contract data;
review maturity and counterparty mapping after product changes;
track deposit and wholesale-funding migration;
update encumbrance states continuously;
stress planned growth before committing the balance sheet;
compare NSFR movements with FTP and liquidity-risk signals;
investigate unusual quarter-end changes;
recalculate after material acquisitions, securitisations or funding programmes.

Research anchors

The deeper lesson

NSFR is the mathematics of refusing to let maturity transformation become invisible. Assets say how long the bank’s money is tied up. Liabilities say how long funding is likely to stay. ASF and RSF translate those properties into weighted quantities and force the two sides to meet. A strong bank therefore does not ask only whether it can fund today’s balance sheet. It asks whether the funding structure deserves to still be there when the next year becomes difficult.

Educational note: This article explains public banking and regulatory-liquidity mathematics. It is not treasury advice, investment advice or an NSFR calculation for any specific institution.

Small Group Tutorials

How Banks Calculate the Net Stable Funding Ratio: ASF, RSF, Funding Tenor, Asset Liquidity and Structural Funding