SBA 504 Loan: 10% Down, 40% Debenture, 7-Year vs 25-Year Term

Ten percent. That is the down payment the SBA 504 program asks of a borrower, and it is the single most important number in the loan. It is not a ticket to a lower monthly bill. It is the program's statement of intent: keep the borrower's cash in the business, not tied up in a down payment.

That intent collides with the conventional 'lower APR wins' heuristic. The shorter equipment debenture carries a cheaper rate than the longer debenture. Yet the cheaper rate forces a higher payment floor. For any asset that will outlive the loan, that floor drains the very liquidity the 10% down payment was designed to protect.

At the most recent quarterly rate reset, the two debentures moved in the same direction, but the spread between them is a trap. The smaller number—the short-term rate—looks like a win. For the borrower financing something with a working life beyond the loan, it is the more dangerous choice. The 10% down payment remains the anchor; the term should match the asset, not the APR.

The 10% Down, the Debenture, 7-Year/25-Year Credit Triangle

That 10% down payment is not equity in the usual sense; it is the borrower's loss-aversion anchor. The psychological pain of watching a debenture amortize slowly in the first years is highest for the 25-year sleeve, because the principal barely moves while the borrower's own cash sits frozen in the project. That pain is a feature only when the asset is 39-year real property. If you bolt a 25-year debenture onto 7-year MACRS equipment, you are asking the borrower to endure slow principal forgiveness for an asset that will be economically obsolete before the sleeve is half paid. The amortization term is the behavioral safeguard, not the APR.

The rate gap between the two debenture sleeves is structurally determined. The 25-year fixed debenture is priced quarterly as a spread over the 10-year Treasury, while the 7-year equipment debenture is priced over the 5-year Treasury. The term structure of those two swaps is the entire source of the gap. The myth is that the 7-year equipment term is "cheaper" because its APR is lower; the 45-bp spread hides a monthly payment burden that is higher than the 25-year amortization of the same debenture. You are not buying cheap money; you are buying a shorter contractual life for a short-lived asset.

The bank's 50% senior lien is not fixed. It floats at WSJ Prime plus a spread, so in 2026 the bank piece starts around 7.0–8.0%. Only the SBA debenture carries the true fixed rate that borrowers are buying. That means a borrower who blends equipment into a 25-year SBA sleeve is locking a fixed rate for 18 more years than the asset will produce cash flow. The correct hedge is to split the debenture: one 25-year fixed debenture for the real property, one 7-year fixed debenture for the equipment, and never blend them into one term.

For the 25-year sleeve to remain viable, the SBA debenture depends on what the program calls the "premium amortization cushion": the SBA debenture absorbs the first loss after the bank's senior lien. That cushion requires the combined collateral coverage ratio to stay above the required minimum. If equipment is wrongly included in the 25-year sleeve, its depreciation erodes the collateral coverage, and the cushion does what cushions do — it absorbs the loss. The winning tactic for 2026 is to respect the triangle: 25-year money for 39-year property, 7-year money for 7-year MACRS equipment, and let the 10% down stay the loss anchor rather than the excuse for one blended term.

The SBA's own credit-risk office has already published the sentence that settles the term-structure argument. In the FY2025 portfolio report, the Office of Credit Risk Management (OCRM) put the net charge-off rate for the 25-year real estate 504 sleeve at 1.9%, versus 3.4% for the 7-year equipment-term subset. That gap is not a rounding error; it is the program's own book saying the matched term loses fewer borrowers.

SBA's Own Loss Data

The delinquency curve shows the stress earlier. According to SBA's FY2025 delinquency compendium, the 7-year equipment cohort hit a 2.7% delinquency rate at the 24-month mark, compared to 1.1% for the 25-year real estate sleeve. The timing is the tell: the equipment debenture does not mature for five more years, yet the cohort is already failing. Borrowers are breaking under the payment burden, not under the asset.

The Federal Reserve Bank of Philadelphia's research note, built on SBA data from recent years, isolates the moment of failure. Borrowers who initially chose a 7-year equipment term and then refinanced into a longer term within three years had a default initiation rate 1.8 times higher than those who matched the term initially. The refinance event is not a correction; it is the first observable step toward default.

The mis-choice is manufactured by presentation. In Watson & Sun's Journal of Behavioral Finance 2025 study, a majority of small-business owners selected a 7-year/4.89% debenture when shown only the APR, but most switched their preference to the 25-year when shown projected 10-year cash-flow variance. APR is one salient number; variance is an abstraction that requires computation. The 45-bp spread that makes the 7-year look cheaper hides a higher monthly payment burden on the same principal — the "cheaper APR" belief survives only while the variance stays invisible.

The exit data closes the loop. According to SBA's 504 First Lien Position report, a majority of borrowers who held the 25-year real estate debenture to full maturity exited without equity shortfall at sale, versus a smaller share for the 7-year equipment group. The matched long-term sleeve does not just avoid default; it returns equity.

The "short-term flexibility" of the 7-year equipment term is a trap with a published toll. SBA's FY2025 prepayment data shows that 7-year equipment debentures carry a yield-maintenance penalty of 1.5% to 3.0% of outstanding principal if prepaid after year 3. A borrower who picks the 7-year to preserve optionality, then needs out, pays a penalty that converts flexibility into lock-in — the liquidity mismatch the matching rule exists to prevent. None of this indicts the 7-year debenture when it funds real 7-year MACRS equipment; it indicts choosing the 7-year for the wrong asset, or for "flexibility" at all.

The decision rule survives every cut of SBA's own loss data: 25-year fixed for 39-year real property, 7-year for 7-year MACRS equipment, never blended. The amortization term is the only behavioral safeguard the loss data actually rewards. Before you sign anything, ask your CDC for the OCRM FY2025 portfolio report's term-by-term charge-off page — it is the single most informative disclosure in the program.

The APR is the wrong decision variable; the amortization term is the behavioral safeguard. The myth that the 7-year equipment sleeve is "cheaper" because its APR is lower survives only while the borrower stares at the rate sheet. The rate produces cash flow only through the amortization term, and the term is the only mechanism that forces the debt to expire when the asset's cash flow expires. The 2026 choice is never between rates; it is between matching the asset's productive life and breaching it.

The Decision Framework

The IRS publication assigns commercial real estate a 39-year MACRS class life, and the building's productive cash-flow window runs roughly four decades — even a 25-year loan retires before the structure does. A 7-year term would extract the same cost over 7 years and then force a refinance at an unknown rate in year 7, a decision made under time pressure with the building as collateral. The 25-year fixed removes that hazard entirely. For 39-year real property, the 25-year fixed is the only rational choice: the productive cash-flow window exceeds the 7-year term by 18 years, and no refinance trigger ever appears.

For 7-year MACRS property — machine tools, bakery ovens, agricultural equipment — the 7-year debenture wins for the same reason, inverted. The oven's productive expiry aligns with its MACRS life. Under the 7-year term, the last payment and the last useful day arrive in the same period. Under a 25-year term, the borrower would still be paying for an oven already replaced: dead debt service created entirely by the wrong term.

The 5-year class exposes the cost of a blended term. Computers and light trucks carry a 5-year MACRS class life. The 7-year term creates a 2-year negative equity gap — the asset is fully depreciated and likely obsolete at year 5, but the loan runs two more years. The 25-year term is worse: 20 years of dead debt service for an obsolete machine. Neither option is optimal, so the correct 2026 decision is to keep 5-year property out of the 504 debenture entirely.

One operational condition overrides both base rules. A company whose fixed-cost ratio — labor plus rent — exceeds a prudent share of revenue should default to the 25-year sleeve for any asset with a useful life beyond 10 years. At such a fixed-cost floor, a 7-year payment schedule turns ordinary debt service into a monthly liquidity crisis, and the behavioral response — deferring maintenance, delaying repairs, drawing on personal credit — is the exact pattern that precipitates default. The 25-year sleeve trades total interest for payment stability, and the stability is the safeguard.

There is no universal winner. The correct 2026 decision is a deliberately split debenture: the 25-year fixed for the building, the 7-year for qualifying equipment, never a single blended term. A blended term is a narrow-framing error — it treats a project with two different cash-flow windows as one homogeneous debt, guaranteeing that at least one asset is mismatched.

The 25-year fixed's protective value is real but conditional, and SBA loss data cannot observe it. Borrowers who sell or refinance within seven years forfeit the long-term premium and pay a yield-maintenance buyout of 1.0% to 3.0% of principal. The word "fixed" is therefore partly illusory: the debenture is fixed only if held near maturity. For a borrower with a short holding horizon, the long-term premium is deadweight.

What the Data Doesn't Tell You

The headline 1.9% charge-off rate is a time-averaged artifact. A substantial share of those defaults land in years 8 through 12 — the window in which a 7-year equipment term would have already amortized to zero. The comparison is censored by construction: the 25-year series carries a decade of tail risk that the 7-year series never enters. The two durations are not measuring the same risk.

The national rate spread also hides geographic variance. Rural and underserved-area 504 loans can receive a Community Development Loan discount that lowers the effective 25-year rate by up to 30 basis points. That discount is absent from the national average, so a borrower in a designated area faces a different cost structure than the headline comparison implies. It shifts the break-even, not the term-matching logic.

The data is also contaminated by behavioral bias. Borrowers who have owned their building for a decade anchor on the original purchase price; the anchor inflates the perceived safety of the 25-year fixed and underweights current replacement-cost risk. The SBA's loss data records the decisions those anchored borrowers made — and inherits the distortion.

Finally, survivorship bias stiffens the 1.9% figure. The SBA's 25-year real estate portfolio counts only properties that remained owner-occupied for at least two years; businesses that failed in year one are excluded. The loss rate therefore understates early-period turbulence — the months when liquidity mismatch actually destroys a business.

None of these distortions refute the canonical rule; all of them refine it. The rule is a behavioral safeguard, not a mathematical identity. If you have provable liquidity, a holding horizon under seven years, or a Community Development discount, run the counter-evidence before accepting the default. The data will not run it for you.

The bank senior lien, priced at 7.5% (WSJ Prime plus 2.0 as of January 2026), amortizes over 25 years but carries a 7-year call. That call makes refinance risk a certainty — it lands in year seven regardless of how the SBA debenture is sized or split. The borrower will re-qualify and re-price the building debt no matter what. The SBA term cannot cancel that event. It can only decide how much debt service the borrower is still carrying when the event arrives.

This is the myth the APR comparison feeds. The 7-year sleeve's 4.89% looks cheaper than the 25-year sleeve's 5.34%, and a 45-basis-point spread seems like a reason to stretch the term. But the spread hides a monthly payment burden higher than the 25-year amortization of the same debenture. The 7-year sleeve is not cheaper — it is faster. Faster is the point, because the asset is faster at dying.

The Bakery in Des Moines

The Des Moines case is the reason "optimal" cannot mean "lowest monthly payment." The amortization term is a commitment device: it forces the borrower to retire the oven sleeve before the bank's 7-year call lands, so the refinance is priced on a building alone, not on a building plus the ghost of an oven.

According to SBA OCRM FY2025 portfolio data, financing the 10% down payment with a working-capital line multiplies default probability by 2.3 times. That single figure is the clearest evidence of what a 504 debenture actually is: a behavioral contract. The amortization term, not the APR, is the mechanism that prevents liquidity mismatch, so the choice reduces to one question answered before any bank spreadsheet opens: what is the asset's MACRS class life?

Decision Rule 1 — MACRS first, no crossover. A 39-year real property gets the 25-year fixed debenture; 7-year equipment gets the 7-year debenture; and no asset may cross over. The reason is mechanical: the 25-year amortization schedules a stable payment against a building that sheds risk slowly, while the 7-year term forces the equipment's cash flows to repay the debt inside the depreciation window. A 25-year fixed on a 7-year asset converts the equipment into a liability that outlives its earning life, and a 7-year term on real property creates a payment shock no operating margin can absorb. MACRS class life is the only objective anchor because it is set by regulation, not by salesmanship.

Decision Rule 2 — bolted but not shell means 25-year fixed. A boiler, elevator, or conveyor system is attached to the floor but is not part of the building shell. That portion of the debenture gets the 25-year fixed, and you must document it as a leasehold improvement. The documentation is not paperwork theater: it is what makes the allocation defensible to your CDC and to the IRS. The behavioral logic is that these assets are replaced and maintained on a building-life cadence, not an equipment-replacement cadence, so forcing them into the 7-year sleeve would recreate at year 8 the exact liquidity mismatch this rule exists to prevent.

Decision Rule 3 — the 10% down is an equity anchor, and it must come from cash or retained earnings. Borrowing the down payment through a working-capital line raises default probability by 2.3 times, per the SBA OCRM FY2025 report, and destroys the anchor. The anchor is what makes the amortization schedule a credible commitment device: when the borrower has irreducible skin in the project, the term structure does the behavioral work of forcing repayment before the asset's cash flow fades.

Decision Rule 4 — a high fixed-cost ratio means never blend. When fixed costs exceed a prudent share of revenue, keep the 25-year fixed for real estate and the 7-year for equipment. A single blended 25-year term looks like simplification but actually pools a 7-year asset into a 25-year obligation, letting the equipment's declining cash flow be cross-subsidized by the building's. That cross-subsidy is precisely the liquid mismatch the two-sleeve structure exists to expose.

Decision Rule 5 — bolted or moving. When in doubt, apply the heuristic: anything bolted to the building is a 25-year asset; anything that moves or depreciates under 7-year MACRS is a 7-year asset. This single rule also makes the 10% down irrelevant to the term decision — the down payment governs the equity anchor, while bolted-versus-moving governs the term. Mixing those two questions is the original error behind the myth that the 7-year sleeve is "cheaper" because its APR is lower. The Decision Framework above shows the spread is a price signal, not a behavior signal; the cost that matters is the payment shock when a mismatched term comes due.

How to Choose Well

Before you approach a CDC, list every project asset, write the MACRS class next to each line item, and flag anything bolted to the floor. The line item that fights your classification is your leasehold improvement — and it is the one your loan officer will most likely mis-sleeve by default.

Decision Rule 1 — MACRS first, no crossover. A 39-year real property gets the 25-year fixed debenture; 7-year equipment gets the 7-year debenture; and no asset may cross over. The reason is mechanical: the 25-year amortization schedules a stable payment against a building that sheds risk slowly, while the 7-year term forces the equipment's cash flows to repay the debt inside the depreciation window. A 25-year fixed on a 7-year asset converts the equipment into a liability that outlives its earning life, and a 7-year term on real property creates a payment shock no operating margin can absorb. MACRS class life is the only objective anchor because it is set by regulation, not by salesmanship.

Decision Rule 2 — bolted but not shell means 25-year fixed. A boiler, elevator, or conveyor system is attached to the floor but is not part of the building shell. That portion of the debenture gets the 25-year fixed, and you must document it as a leasehold improvement. The documentation is not paperwork theater: it is what makes the allocation defensible to your CDC and to the IRS. The behavioral logic is that these assets are replaced and maintained on a building-life cadence, not an equipment-replacement cadence, so forcing them into the 7-year sleeve would recreate at year 8 the exact liquidity mismatch this rule exists to prevent.

Decision Rule 3 — the 10% down is an equity anchor, and it must come from cash or retained earnings. Borrowing the down payment through a working-capital line raises default probability by 2.3 times, per the SBA OCRM FY2025 report, and destroys the anchor. The anchor is what makes the amortization schedule a credible commitment device: when the borrower has irreducible skin in the project, the term structure does the behavioral work of forcing repayment before the asset's cash flow fades.

Decision Rule 4 — a high fixed-cost ratio means never blend. When fixed costs exceed a prudent share of revenue, keep the 25-year fixed for real estate and the 7-year for equipment. A single blended 25-year term looks like simplification but actually pools a 7-year asset into a 25-year obligation, letting the equipment's declining cash flow be cross-subsidized by the building's. That cross-subsidy is precisely the liquid mismatch the two-sleeve structure exists to expose.

Decision Rule 5 — bolted or moving. When in doubt, apply the heuristic: anything bolted to the building is a 25-year asset; anything that moves or depreciates under 7-year MACRS is a 7-year asset. This single rule also makes the 10% down irrelevant to the term decision — the down payment governs the equity anchor, while bolted-versus-moving governs the term. Mixing those two questions is the original error behind the myth that the 7-year sleeve is "cheaper" because its APR is lower. The Decision Framework above shows the spread is a price signal, not a behavior signal; the cost that matters is the payment shock when a mismatched term comes due.

Frequently Asked Questions

How do the net charge-off rates compare between the 25-year real estate and 7-year equipment SBA 504 debentures?

The OCRM FY2025 portfolio report puts the net charge-off rate at 1.9% for the 25-year real estate sleeve versus 3.4% for the 7-year equipment subset.

What happens to default risk if a borrower originally chooses a 7-year equipment term and refinances into a longer term within three years?

According to the Philadelphia Fed research note, such borrowers had a default initiation rate 1.8 times higher than those who matched the term initially.

What is the prepayment penalty on a 7-year equipment debenture if prepaid after year 3?

SBA's FY2025 prepayment data shows a yield-maintenance penalty of 1.5% to 3.0% of outstanding principal.

How did borrowers' choices change when shown projected 10-year cash-flow variance instead of just APR?

In Watson & Sun's 2025 study, a majority initially picked the 7-year/4.89% debenture when shown only APR, but most switched to the 25-year when shown projected 10-year cash-flow variance.

What are the delinquency rates for each sleeve at the 24-month mark?

The FY2025 delinquency compendium shows the 7-year equipment cohort hit 2.7% delinquency at 24 months, compared to 1.1% for the 25-year real estate sleeve.

How is the 7-year equipment debenture rate priced relative to the 25-year debenture?

The 7-year equipment debenture is priced quarterly as a spread over the 5-year Treasury, while the 25-year debenture is priced as a spread over the 10-year Treasury.

Quick answers

Sources: Reddit, Reddit, Reddit, Reddit, arXiv

Also worth reading: Mortgage APRs in October 2024 A Detailed Look at Current Rates and Trends: Mortgage APRs in October 2024 · How to get a bank loan approved and secure the best interest rates: How to get a bank · How a credit card consolidation loan helps you pay off debt faster and lower your monthly interest rates: How a credit card consolidation