Ask a restaurant owner why a loan request was declined and you will usually hear something about credit, or the industry, or a slow quarter. Ask the analyst who worked the file and you will often hear a single number: occupancy cost. Rent and its associated charges are the one large expense in a restaurant that is fixed for a decade, cannot be re-negotiated when sales soften, and sits ahead of a lender in every practical sense. A landlord can evict. A lender holding a lien on used fryers cannot do much of anything.
That asymmetry is why occupancy cost gets computed before anything else. It is a fast calculation, it is hard to argue with, and it explains most of the variation in restaurant credit outcomes. Two concepts with identical sales, identical food cost and identical owners will be underwritten very differently if one signed at 7 percent of sales and the other at 12. This article covers how the ratio is built, what it has to clear, why the lease term matters as much as the rent, and what an operator can actually do when the number comes in high.
What occupancy cost actually includes
Occupancy cost is not base rent. It is every dollar you pay to be in the building, and operators routinely understate it by a third because they only count the number on the first page of the lease.
- Base rent, including any scheduled escalations already in force.
- Common area maintenance, whether billed monthly or reconciled annually.
- Real estate taxes passed through to the tenant.
- Building and liability insurance the lease requires you to carry or reimburse.
- Percentage rent: the additional rent owed above a stated sales breakpoint.
- Marketing or promotional fund contributions, common in centers and malls.
- Trash, grease, and pest charges billed by the landlord rather than a vendor.
Add those, divide by net sales: sales after discounts, comps and sales tax, not gross rung on the register, and you have the ratio. Use net sales consistently. Dividing by gross receipts flatters the number and any analyst will re-run it the other way.
The bands, and how a desk reads them
There is no universal cutoff, because the workable range depends on service model, market and volume. A quick-service unit doing high revenue out of a small footprint tolerates a higher rent per square foot than a full-service dining room. What is consistent is the shape of the reaction as the ratio climbs.
| Occupancy cost as a share of net sales | How it usually reads | Practical effect on the file |
|---|---|---|
| Under 6% | Favorable lease, or strong volume per square foot | Rarely the issue; attention moves elsewhere |
| 6% to 8% | The normal working range for full service | Neutral; the file turns on prime cost |
| 8% to 10% | Tight but common in high-rent trade areas | Sizing gets conservative |
| 10% to 12% | The lease is consuming the margin that would service debt | Often the binding constraint |
| Above 12% | Structurally difficult without volume growth | Frequently declined on cash flow |
Nothing in that table is a commitment, and no ratio produces an automatic outcome in either direction. Treat the bands as the shape of the conversation you are going to have.
Why occupancy sits above everything else in the stack
Restaurant cost structure is usually described through prime cost, food and beverage plus total labor. Prime cost is the number a good operator manages weekly, and it is genuinely where the craft is. But prime cost is variable. It moves with volume, with menu engineering, with scheduling discipline. Occupancy does not move at all.
| Cost line | Illustrative range as a share of net sales | How much control the operator has |
|---|---|---|
| Food and beverage cost | 28% to 34% | High: purchasing, portioning, menu mix |
| Labor, including taxes and benefits | 30% to 36% | High: scheduling, station design, turnover |
| Prime cost (the two combined) | 60% to 67% | The number a good operator manages weekly |
| Occupancy | 6% to 10% | Effectively none once the lease is signed |
| Other operating expense | 12% to 16% | Moderate: utilities, repairs, supplies, marketing |
| Left for debt service, taxes and the owner | 3% to 10% | Whatever survives the lines above |
Read the last two rows together and the underwriting logic becomes obvious. If prime cost runs 65 and other operating runs 14, that is 79 percent gone. Occupancy at 7 leaves 14 points for debt service and profit. Occupancy at 12 leaves 9. On $1.45 million of sales that difference is roughly $72,000 a year: for most restaurant files, the entire debt service capacity.
Lease term is collateral term
The second thing the desk pulls from the lease has nothing to do with rent. It is the remaining term. A leased restaurant offers very little hard collateral: the buildout is attached to somebody else's building, the hood and walk-in are fixtures, and the resale market for used kitchen equipment is thin. What the lender is really lending against is the right to keep operating in that location for the life of the loan.
So the remaining term (usually including renewal options the tenant controls unilaterally) is generally expected to meet or exceed the term of the loan. A request for ten-year money against four years of remaining lease creates an obvious gap, and the usual fixes are a lease extension negotiated before the file goes in, or a shorter amortization that costs more per month.
Escalators, percentage rent and the ten-year view
Occupancy cost is not one number, it is a curve. A 3 percent annual escalator compounds: $13,400 a month becomes about $15,500 by year five and roughly $18,000 by year ten. If sales are flat over that window (which happens) an 8 percent occupancy ratio drifts toward 11. Underwriters who work restaurants regularly model that forward, because the loan has to survive the later years, not the first one.
Percentage rent adds a second layer. A typical structure charges additional rent on sales above a breakpoint, which means the landlord participates in exactly the growth you were counting on to fix the ratio. It is not automatically bad (breakpoints are often set high enough to be irrelevant) but it should be modeled, not assumed away.
What to do when the ratio is too high
There are only three levers, and only one of them is fast. Raising sales fixes the denominator but takes quarters. Reducing occupancy fixes the numerator and can happen in weeks if the landlord has a reason to engage. Changing the footprint: subleasing dead square footage, giving back a patio you never staffed: does both.
Before you conclude the lease cannot move
- Compute the actual ratio using trailing twelve month net sales and the last full CAM reconciliation, not the monthly estimate.
- Calculate sales per square foot and compare it to what the space costs per square foot, the two numbers together tell you whether you are overpaying or underperforming.
- Check the escalation schedule and model occupancy in year five and year ten at flat sales.
- Confirm remaining term including only options you can exercise without consent.
- Read the assignment and subletting clauses before you assume you are stuck with the whole footprint.
- Price a blend-and-extend: lower rent now in exchange for term, which frequently helps the lender as much as the cash saving does.
- Get any negotiated change into a signed amendment before the file is submitted. Verbal accommodations are not underwritable.
How the lease actually gets read in the file
Expect the whole document, not a summary, and expect specific provisions to be pulled out. Preparing a one-page lease abstract yourself: commencement, expiration, options and their exercise mechanics, base rent schedule, NNN structure, percentage rent breakpoint, assignment terms, personal guarantee and any burn-off, removes days from the process and signals that you know what you signed.
Is occupancy cost measured against gross sales or net sales?
Net sales: after discounts, comps, employee meals and sales tax. Using gross receipts understates the ratio by whatever your discounting runs, and an analyst will recompute it on net anyway. Be consistent between the P and L, the POS reports and anything you present.
My rent is high but my volume is strong. Does that fix it?
Usually yes, because the ratio is what matters, not the rent per square foot. A unit paying $65 a foot on sales of $900 a foot is in better shape than one paying $32 on sales of $260. That is exactly why the metric is expressed as a percentage of sales rather than an absolute number.
Does owning the building instead of leasing change the analysis?
It changes the shape of it. Owner-occupied real estate replaces rent with mortgage debt service, which is fixed but builds equity and often lands lower than market rent. The desk will still compute the same ratio using the actual debt service, and will also underwrite the property itself. Financing structures for owner-occupied space are different from leasehold financing, and generally longer.
How do lenders treat a percentage rent clause?
As a variable cost layered on top of base occupancy, modeled at the sales level being projected. If the natural breakpoint is well above realistic volume it may be treated as immaterial. If your projections cross the breakpoint, expect the additional rent to be added to occupancy in the model, which is the correct treatment.
I have a personal guarantee on the lease. Does that affect a loan request?
It can. A lease guarantee is a contingent liability, and on a ten-year lease at $160,000 a year the exposure is not small. It belongs on your personal financial statement. Many restaurant leases include a burn-off after a period of on-time payment: if yours does, note it, because the remaining exposure is what matters.
Can I get the lease renegotiated after I have already signed?
Sometimes, particularly if the landlord has vacancy in the center, if you are a strong tenant with a long track record, or if you can offer term in exchange. It is a commercial conversation, not a hardship appeal. Come with your sales per square foot, comparable asking rents in the trade area and a specific proposal.
Will a lender lend on a location with under two years of lease left?
Rarely on a term of any length, because the collateral position ends when the lease does. The usual sequence is to extend the lease first, then apply. If the extension is genuinely unavailable, the realistic options are shorter amortization, a smaller request, or capital secured by something other than the location.
Where to start
Compute the ratio this week, using trailing twelve month net sales and the last full CAM reconciliation. Then build the one-page lease abstract. Those two artifacts take an afternoon and they will tell you, before anyone else does, whether your file turns on the lease or on the kitchen.
If the number lands above the range and the lease has term left, work the landlord conversation before the funding conversation. If the number is fine and the constraint is somewhere else, you have just saved yourself from fixing the wrong thing.
Qualified Commercial Underwriting Desk
Credit and capital markets
The Qualified Commercial underwriting desk reviews commercial real estate, dealer and Main Street files daily. The Academy is written from that work (how files are actually read, priced and declined) rather than from a rate sheet.
Educational content only. Nothing here is a commitment to lend, an offer of credit, or tax, legal or accounting advice. Program terms, timelines and thresholds vary by lender, file and market conditions, and any figures shown are illustrative.