Loan-to-Deposit Ratio as a Liquidity and Revenue Metric for Credit Unions in 2027

The Tuesday morning that starts every liquidity problem

Picture a $500 million credit union heading into a routine ALCO meeting. The treasurer pulls the month-end board packet: loan-to-deposit ratio 79%, comfortably inside policy. Nobody flags anything. Three weeks later the CFO runs an ad-hoc query because a member service manager mentioned an unusual number of large share withdrawals, and the ratio comes back at 88%. Nothing dramatic happened. A commercial member drew a big chunk of a previously unused line of credit, indirect auto volume ran hot for a month because a dealer partner had a strong quarter, and a handful of rate-sensitive members moved money-market balances to a competitor advertising a promotional certificate. Three ordinary events, none of which triggered an alert, stacked inside a single reporting gap.

That is the whole problem in miniature. The loan-to-deposit ratio is a stock measure — a photograph of the balance sheet at one instant — but the forces that move it are flows: origination volume, line draws, share inflows, certificate maturities, payroll deposit timing. When you observe a stock quantity monthly while the underlying flows run daily, you are structurally blind to the inflection point. You see the ratio after it moved, not while it was moving, and by then your options have narrowed from "adjust pricing gently" to "stop lending."

The narrowing matters more for a credit union than for a comparably sized community bank, and this is the part that gets underweighted in generic asset-liability writing. A bank facing the same squeeze has a wider menu: it can issue brokered certificates, tap the Federal Home Loan Bank system against pledged collateral, raise wholesale funds, or in some cases go to the capital markets. A credit union is a member-owned cooperative with no outside equity holders. Its capital comes almost entirely from retained earnings, and its funding comes almost entirely from member shares. Contingent sources exist — FHLB advances against eligible collateral, the NCUA's Central Liquidity Facility for members that have joined and paid in, corporate credit union lines, and the Federal Reserve's discount window — but each requires pre-positioned collateral, prior membership, and operational readiness that must be established *before* the squeeze, not during it.

That asymmetry is why the ratio functions as a genuine constraint rather than a dashboard decoration. If a credit union lets the ratio run to the low nineties without contingent funding tested and available, its remaining levers are slow and expensive: raise share rates enough to pull deposits back (which compresses margin immediately across the entire existing deposit book, not just new money), price loans up to slow origination (which takes a full pipeline cycle to bite), or sell or participate out loans (which takes weeks and typically realizes a discount in a rising-rate environment). None of those is a same-week fix. The fix has to be installed before you need it.

Frame the scenario correctly and the operating question changes shape. It is not "what is our ratio?" It is "what will our ratio be in ninety days given committed-but-unfunded loans, undrawn commercial lines, scheduled certificate maturities, and observed share behavior — and which lever do we pull today so that number lands inside policy?" That forward-looking version is what a competent ALCO actually manages, and it is why the reporting cadence matters as much as the threshold itself.

How the mechanism actually works

Start with the arithmetic, because the definitional sloppiness here causes real disagreements between the finance team and the board. The core calculation is total loans outstanding divided by total shares and deposits, expressed as a percentage. But institutions differ on the numerator and denominator in ways that shift the result by several points:

Numerator choices. Do you use gross loans or net of the allowance for credit losses? Do you include loans held for sale? Do you include loan participations you have purchased (you should — you own the credit and it consumes liquidity) and exclude participations you have sold without recourse (also correct — the balance left)? Do you count the unfunded portion of committed lines? Standard practice is no for the ratio itself, but yes in a separate forward-exposure view, because undrawn commercial and home-equity lines are exactly what surprised the credit union in the scenario above.

Denominator choices. Member shares of all types — regular shares, share drafts, money market shares, share certificates, IRA shares — belong in. Non-member deposits and public unit deposits, where permitted, are often shown both ways because they behave differently under stress. Borrowings do not belong in the denominator of a *loan-to-deposit* ratio; if you fold FHLB advances in, you have built a loan-to-funding ratio, which is a legitimate metric but a different one, and it will read materially lower. Pick one definition, write it into ALM policy, and report the other as a supplemental line so nobody compares across incompatible bases.

The mechanism that moves the ratio has two independent drivers, and separating them is the analytical heart of the whole exercise. The ratio can rise because loans grew, or because shares shrank, and the correct response is opposite in each case. Loan-driven increases respond to loan pricing, underwriting tightening, participation sales, and origination target adjustments. Share-driven increases respond to deposit pricing, retention campaigns, and channel promotion. Treating a share-runoff problem with a lending brake wastes revenue you did not have to give up; treating a loan-growth problem with a certificate special buys expensive money you did not need.

The sensitivity math is worth internalizing because it converts a percentage into decisions. On a balance sheet with $400 million in loans and $500 million in shares, the ratio is 80%. Fund $10 million of net new loans with no share change and the ratio moves to 82%. Lose $10 million of shares with no loan change and it moves to roughly 81.6%. So on a book that size, every $10 million of net loan growth costs about two points of headroom, and every $10 million of share runoff costs about one and a half. If policy caps you at 85%, you have roughly $25–30 million of combined room. Say that to a lending team and it lands very differently from "keep an eye on liquidity."

The second mechanical point is that the ratio is joined at the hip to net interest margin, and optimizing either one alone produces a bad outcome. Loans yield more than the alternatives a credit union has for parking member money — overnight balances, agency securities, short Treasuries — so moving deposits from investments into loans raises interest income. That is the revenue argument for a higher ratio. But the marginal loan added at the top of a growth push is usually the least attractive one: thinner spread, weaker credit, or won on price rather than relationship. And the marginal deposit needed to fund it is the most expensive one, because rate-shopping money comes in at the top of the market and often reprices the existing book alongside it. Push the ratio up hard enough and margin can fall even as the ratio rises — the exact opposite of the intended effect.

That is why the sensible target is a band, not a maximum. A band says: inside this range the marginal loan and the marginal deposit are both priced sensibly, so grow. Outside it in either direction, something in the pricing or the mix needs attention before volume does.

Real numbers, ranges, and benchmarks that hold up

Concrete figures make this operational, so here are the ones worth anchoring on — with an honest note on which are industry conventions versus which you must pull fresh.

The working band. For a typical retail-focused credit union, 70–80% is the conventional comfort zone, with 65–85% a defensible wider policy band depending on funding profile and contingent liquidity readiness. Below about 60%, earnings drag becomes the dominant concern. Above about 90%, liquidity risk and supervisory attention become the dominant concern. These are practitioner conventions, not statutory thresholds — there is no regulation that says "90% is illegal." What exists is a supervisory framework in which the liquidity component of the CAMELS rating considers funding concentration, contingent sources, and the loans-to-shares relationship together, and a sustained high ratio without demonstrated contingent capacity is exactly the pattern that draws an examiner's questions.

Peer benchmarking, done properly. The NCUA publishes quarterly Call Report data and financial performance reports for every federally insured credit union, free. Use it — but use it with the right peer set. Comparing a $150 million single-sponsor credit union to a $4 billion multi-common-bond institution produces noise. Build a peer group of roughly ten to twenty institutions matched on asset size band, field of membership type, and geography, and pull their loans-to-shares figures each quarter. Look at the distribution, not just the median: if your peer set runs 68% to 84% and you sit at 91%, you are an outlier and should be able to explain why in one sentence. Because these figures move every quarter, pull the current number rather than citing a remembered one.

The revenue arithmetic. This is where the ratio earns its place as a revenue metric, not just a liquidity metric. Take a $500 million share base. Ten points of ratio is $50 million of balances. If those dollars sit in short-term investments versus being deployed into a loan portfolio, the difference in yield is the spread between the two — and that spread varies enormously with the rate environment. In a flat or inverted curve, short investments may yield close to consumer loan rates net of expected credit losses, and the earnings case for pushing the ratio up weakens considerably. In a steeper curve with wide consumer loan spreads, the same ten points might be worth well over a million dollars of annual interest income before credit costs. The discipline is to compute this with *your current* marginal loan yield, *your current* investment yield, and *your current* expected loss rate — not a rule of thumb. The formula is simple: incremental balances × (marginal loan yield − expected credit loss rate − investment alternative yield − incremental servicing cost). Run it every quarter. The answer changes.

Growth-rate leading indicators. Because the ratio is a stock, watch the flows that drive it. If annualized loan growth is running more than roughly eight to ten percentage points above share growth, the ratio is climbing at a pace that will consume a typical policy band inside a year. That gap — loan growth minus share growth — is arguably a better weekly dashboard number than the ratio itself, because it moves first.

Deposit beta. Beta is the share of a market rate move you pass through to members: change in your deposit rate divided by change in the benchmark rate. Low beta protects margin in the short run and starves share growth in the medium run. High beta defends balances and compresses margin. Credit unions have historically run somewhat lower betas than comparable banks, reflecting relationship stickiness, but betas rose across the industry through the recent tightening cycle as members became more rate-aware and digital account opening lowered switching costs. Model beta by product — share drafts, regular shares, money market, and certificates behave nothing alike — and by member segment. A single blended beta hides the fact that a small slice of large, rate-sensitive balances drives most of your runoff risk.

Non-maturity deposit behavior. Checking and savings balances have no contractual maturity, which makes them your cheapest and least predictable funding. Estimate decay rates from your own history rather than borrowing an industry assumption, and stress them: what does the ratio do if non-maturity balances run off faster than your base case for two consecutive quarters? Regulators expect that scenario to be modeled and documented, and it is the single most useful stress test for a credit union because non-maturity runoff is precisely how a comfortable ratio becomes an uncomfortable one without any lending decision being made.

Concentration overlay. Two credit unions at 82% can face completely different risk. One funds itself with fifty thousand small retail share accounts. The other has a fifth of its shares in a few dozen large balances, some above the insured limit. The second is far more fragile at the same ratio, because a handful of decisions can move the denominator. Always report the ratio next to your largest-depositor concentration and your uninsured share percentage. The three numbers together tell a story that the ratio alone does not.

Trade-offs, alternatives, and when a different lever is better

Every action that moves the ratio costs something. Naming the cost explicitly turns ALCO discussion from opinion into comparison.

Raising share rates. The fastest way to pull the denominator up. The cost is that repricing is rarely surgical — raise the money market rate and you typically pay it to existing balances as well as new ones, so the marginal cost of funds on new money is much higher than the headline rate suggests. If you raise a rate 50 basis points on a $150 million money market book to attract $15 million of new balances, you pay the increase on $165 million to acquire $15 million. The effective marginal cost is many times the headline. Tiered and new-money-only structures blunt this, at the cost of member-fairness optics that matter more in a cooperative than in a bank. Certificate specials with defined terms are usually the cleaner tool: they buy a known amount of money for a known duration at a known price, and they mature, which restores flexibility.

Raising loan rates. Slows origination and improves spread on what you do book. The cost is slow: pricing changes take a full pipeline cycle to show in funded balances, so this is a steering input, not a brake. It also has competitive consequences — in indirect auto especially, dealers route to whoever is sharpest, and pricing yourself out of a channel can mean losing the relationship rather than just the volume, with a long road back.

Selling or participating loans. Converts loans directly into liquidity and drops the ratio immediately. The costs are execution time, pricing (in a rising-rate environment, older lower-coupon paper sells at a discount), loss of the member relationship if servicing transfers, and the operational overhead of participation agreements. Participations sold with servicing retained are usually the better structure. This is a genuine lever, but it works best when arranged before you are visibly desperate.

Borrowing. FHLB advances against pledged collateral, corporate credit union lines, the NCUA's Central Liquidity Facility for members, and the Federal Reserve's discount window are all real contingent sources. Note that borrowings do not lower the loan-to-deposit ratio as conventionally defined — they add funding without changing the denominator. What they do is address the underlying liquidity need, which is the thing the ratio is a proxy for. This distinction matters: an institution can be at 92% and perfectly liquid because it has substantial tested borrowing capacity, or at 84% and fragile because it has none. Do not let the ratio substitute for an actual liquidity assessment.

Holding a lower ratio deliberately. Some credit unions with very sticky, low-cost, high-volume share bases run in the sixties on purpose. They are accepting lower interest income in exchange for structural resilience and the ability to grow lending opportunistically without funding stress. Given a large non-interest-bearing checking base and a strong non-interest income stream, that is a coherent strategy, not a failure to lend. The failure version is running low because origination capability is weak and nobody has addressed it — same ratio, entirely different diagnosis.

The meta-trade-off is between margin today and optionality tomorrow. A credit union that runs at the top of its band with no tested contingent funding is buying current earnings with future flexibility. That can be the right call in a stable funding environment with a diversified member base. It is a poor call when share concentration is high, when the rate environment is moving, or when a large share of balances sits above the insured limit. Make the choice consciously and write down why, so that the next examiner conversation and the next board conversation both have an answer ready.

Pitfalls that turn a manageable ratio into a directive

Reviewing monthly and calling it monitoring. The single most common failure. A monthly cadence guarantees you learn about a five-point move three weeks after it started. Weekly is the practical minimum for the ratio itself; daily is right for the inputs if the core system supports an automated extract. The cost of a daily extract is a small amount of engineering, and it converts the ratio from a report into a control.

Watching the level and ignoring the slope. An institution at 84% and flat is in a completely different position from one at 79% and climbing two points a month. The second will breach first. Report the level, the trailing thirteen-week trend, and the ninety-day forecast together, every time. A number without a direction is close to useless.

Forgetting undrawn commitments. The line-of-credit draw that surprises the treasurer was never a surprise — it was a commitment already on the books that nobody was tracking as a forward exposure. Maintain a standing view of undrawn commercial lines and home-equity availability, apply a historically grounded utilization assumption, and carry the result in the forward projection. This is the highest-value single addition to most ratio dashboards.

Treating the ratio as the liquidity policy rather than one input to it. A credit union can sit inside its band and still be illiquid: on-balance-sheet liquid assets thin, contingent lines untested, collateral unpledged, no operational rehearsal of a draw. The ratio is a useful summary indicator. It is not a liquidity coverage measure, it says nothing about asset quality, and it says nothing about whether you can actually access funding on a Friday afternoon. Pair it with a liquid-asset measure, a contingency funding plan with named triggers and owners, and a documented annual test of each contingent source.

Chasing the peer median. The median is a description of what other institutions chose, not a recommendation for you. A credit union with high share concentration and heavy uninsured balances should deliberately run below the peer median. One with an exceptionally stable single-sponsor member base can justify running above it. Benchmarking is for detecting outliers worth explaining, not for setting targets by imitation.

Solving a share problem with a lending brake. If the ratio rose because shares ran off, halting originations gives up loan revenue without addressing the cause, and the deposits keep leaving. Always run the driver test — decompose the change into the loan-growth contribution and the share-change contribution — before choosing the lever. It takes about a minute with two data points and prevents the most expensive category of mistake in this whole discipline.

Letting the definition drift. When the ALM model, the board packet, and the CFO's spreadsheet each compute the ratio slightly differently, meetings turn into reconciliation exercises. Write the formula into policy — exact numerator, exact denominator, exact source fields — and have one system of record produce the number everyone quotes.

Never testing contingent funding. Having an FHLB relationship is not the same as having tested a draw, confirmed collateral eligibility, verified pledging is current, and walked the operational steps end to end. Test each contingent source at least annually with a small live transaction, document the result, and note the time from request to funds. That documented time is what determines how much ratio headroom you actually need.

Building the dashboard and not wiring the triggers. A dashboard nobody acts on is decoration. Attach named thresholds to named owners with named actions: at this level the treasurer reviews deposit pricing; at this level ALCO convenes within five business days; at this level origination targets are formally revisited and the board is notified. Approve the triggers in advance, in calm conditions, so that decisions under pressure are executions of an existing plan rather than improvisations.

Related questions

How is loan-to-deposit different from a liquidity coverage measure?

Loan-to-deposit compares two balance-sheet totals and indicates structural funding pressure. A liquidity coverage measure compares high-quality liquid assets to projected short-horizon outflows. The first says how deployed your balance sheet is; the second says whether you can meet near-term obligations. Report both.

Should undrawn lines of credit count in the numerator?

Not in the headline ratio, which conventionally uses funded balances. But carry undrawn commercial and home-equity availability in a separate forward-exposure view with a historically grounded utilization assumption, because unexpected draws are a leading cause of sudden ratio jumps.

Does a high ratio automatically hurt a CAMELS rating?

No single number determines a rating. The liquidity component considers funding stability, concentration, contingent capacity, and management's monitoring alongside the ratio. A high ratio backed by tested contingent funding and rigorous monitoring reads very differently from a high ratio with neither.

How should a small credit union benchmark without a peer analytics subscription?

NCUA Call Report data and financial performance reports are published quarterly and free. Build a peer set matched on asset band, charter and field-of-membership type, and geography, pull loans-to-shares each quarter, and track your percentile position rather than the raw gap to the median.

What is the right policy band to write into ALM policy?

Set a floor where earnings drag becomes unacceptable and a ceiling where your tested contingent funding can no longer bridge a stress scenario. For many retail credit unions that lands near 65–85%, but derive it from your own runoff modeling and funding capacity, not from a published convention.

FAQ

What loan-to-deposit ratio should a credit union target?

A 70–80% band suits most retail-focused credit unions, with 65–85% a reasonable policy range depending on funding profile. Below roughly 60%, member deposits are underdeployed and interest income suffers. Above roughly 90%, funding pressure and supervisory attention rise sharply. These are practitioner conventions rather than regulatory limits — the right band for your institution should be derived from your own share-runoff modeling and your tested contingent funding capacity.

How often should the ratio be calculated and reviewed?

Weekly at minimum, with daily automated extraction of the underlying loan and share balances if your core system supports it. Monthly review is too slow: the ratio can move five points inside a single reporting gap through a combination of a large line draw, a strong origination month, and ordinary certificate maturities. Report the level, the trailing trend, and a ninety-day forward projection together.

Why does this ratio matter more for credit unions than for banks?

Credit unions are member-owned cooperatives without outside equity holders. Capital comes from retained earnings and funding comes overwhelmingly from member shares. Contingent sources exist — FHLB advances, the Central Liquidity Facility, corporate credit union lines, the discount window — but each requires prior membership, pledged collateral, and operational readiness arranged in advance. That narrower and slower menu makes the ratio a binding operating constraint rather than a monitoring statistic.

Is a low ratio always a problem?

No. Some credit unions run in the sixties deliberately, trading interest income for structural resilience and the ability to grow lending opportunistically. With a large low-cost checking base and solid non-interest income, that is a coherent strategy. The problem case is a low ratio caused by weak origination capability that nobody has addressed — identical number, entirely different diagnosis and remedy.

Can the ratio be used to set loan and deposit pricing?

Yes, and that is its most practical use as a revenue metric. Build a pricing matrix tied to bands: below the floor, sharpen loan pricing to stimulate demand; inside the band, hold; above the ceiling, raise loan pricing and consider a targeted certificate offer. Approve the matrix in advance so pricing decisions under pressure execute an existing plan rather than improvising.

What is the fastest way to bring a high ratio down?

There is no genuinely fast lever, which is the point of monitoring forward rather than backward. Selling or participating loans acts most directly but takes weeks and prices at market. Certificate specials buy a known amount for a known term at a known cost. Loan repricing works but only across a full pipeline cycle. Drawing tested contingent funding addresses the underlying liquidity need immediately, though it does not lower the conventionally defined ratio itself.

Sources

• NCUA Credit Union and Corporate Call Report Data
• NCUA Financial Performance Reports
• NCUA Liquidity and Contingency Funding Plans Rule
• NCUA Central Liquidity Facility
• FFIEC Uniform Financial Institutions Rating System (CAMELS)
• Federal Housing Finance Agency: Federal Home Loan Bank System
• Federal Reserve Discount Window
• FDIC Risk Management Manual of Examination Policies — Liquidity
• Federal Reserve H.8 Assets and Liabilities of Commercial Banks

Related on PULSE

• [Credit Card Interchange Revenue per Active Account: Banking Fee Income](/knowledge/ik0564)
• [Net Revenue Retention (NRR) for SaaS: Churn Mitigation as a Growth Metric](/knowledge/ik0527)
• [Top 10 Insurance Loss Ratio and Premium Revenue Indicators](/knowledge/ik0498)
• [Top 10 Insurance Loss Ratio and Combined Ratio Benchmarks](/knowledge/ik0541)
• [Tuition Revenue per Enrolled Student: Private School Financial Health Metric](/knowledge/ik0553)
• [Hotel Management: RevPAR vs. TRevPAR — Which Metric Drives Profitability Decisions?](/knowledge/ik0585)