What is a base-year stop in an office lease and how does it affect the model?

A base-year stop sets the tenant's opex reimbursement obligation relative to actual opex in a fixed base year, usually the year the lease commenced. The model has to gross up current-year opex to the lease-defined occupancy standard (often 95% or 100%) before computing each tenant's reimbursement, skipping the gross-up misstates the recovery on both sides.

How much should an office model reserve for TI and leasing commissions?

It depends on the WALT and the split between new and renewal leasing, but $1.50–2.50/RSF/year is a reasonable range for a Class A asset with a WALT under five years. Reserving less than that on a rollover-heavy asset overstates distributable levered cash flow.

What is sublease shadow space and why does it matter for underwriting?

Shadow space is square footage a tenant is subleasing, usually below its own contractual rent, while remaining on the master lease. It signals coming vacancy and demand softness before the direct rent roll shows any change, and effective market rent should be haircut against it when modeling re-leasing assumptions.

How does WALT affect the exit cap assumption in an office model?

A shorter WALT at exit means the buyer of the asset is taking on more near-term rollover risk, which typically widens the exit cap relative to a longer-WALT comparable. Modeling a fresh re-lease with an extended WALT can justify exit cap compression, but only if the leasing assumption is grounded in submarket absorption data.

What is a real estate underwriting model?

A structured spreadsheet that projects a property's cash flow, financing, and returns under explicit assumptions — the basis for every institutional acquisition decision.

Office · Underwriting Model

Office Underwriting Model

How to build an office underwriting model: stacking-plan revenue, base-year stop math, TI/LC reserves, sublease shadow space, and exit cap. Free template.

Get the Excel template Automate it with Moraine

Why the underwriting model looks different for office

Office revenue is built from a stacking plan (suite-by-suite rent, expiration, and TI/LC obligations) with base-year stop reimbursement math computed tenant-by-tenant, because each lease's base year and opex gross-up standard can differ from the next.

TI/LC burn belongs in the levered cash flow itself. With WALT often under five years, the model reserves per RSF for both new and renewal leasing at rollover, and that reserve materially changes what a straight NOI number shows.

Sublease shadow space (space a tenant is trying to sublet below its own contractual rent) signals demand softness the direct rent roll doesn't capture, and effective market rent has to be haircut against it before the model can trust its own re-leasing assumptions.

The office-specific checklist

1

Build the rent roll

Verify: unit- or tenant-level detail with rent, term, recoveries
2

Project market rent growth

Verify: submarket-specific; avoid flat nationals assumption
3

Model recoveries explicitly

Verify: CAM, tax, insurance broken out, not blended
4

Run sensitivity on exit cap

Verify: ±75 bps and stress the downside
5

Build debt sizing

Verify: LTV and DSCR constraints both checked
6

Compute leveraged IRR and equity multiple

Verify: cash flow to equity, not enterprise
7

Build a stacking plan by suite with expiration, rent, and TI/LC obligations

Verify: every suite reconciles to total occupied RSF and total in-place rent
8

Compute base-year stop reimbursement tenant-by-tenant

Verify: each tenant's base year and opex gross-up standard checked against its actual lease, not a blanket recovery %
9

Gross up opex to the lease-defined occupancy standard before computing reimbursement

Verify: opex grossed to 95% or 100% per lease language, not left at actual trailing occupancy
10

Quantify sublease shadow space and haircut effective market rent

Verify: shadow space square footage and asking rent gap sourced from current listings, not the rent roll alone
11

Reserve TI/LC per RSF rolling, split new vs. renewal assumptions

Verify: reserve reflects the RSF-weighted mix of leases expiring during the hold, not a flat blended $/RSF
12

Stress-test credit concentration among the top tenants

Verify: quantified NOI impact if the top 3 tenants by GPR vacate simultaneously

Metrics that matter for office

Metric	Target	Calculation
WALT at exit	>4.5 years	Σ(RSF × remaining lease term) / total leased RSF
TI/LC reserve coverage	100% of modeled rollover cost	annual reserve / (RSF rolling × $/RSF TI + LC)
Sublease discount	<15% below direct asking	(direct asking rent − sublease asking rent) / direct asking rent
Net effective rent	≥80% of face rent	(face rent × term - TI - free rent value) / term
WALT (yrs)	>5 for core	weighted avg remaining lease term
Occupancy	>90% for core	leased SF / rentable SF

Red flags unique to office

Reimbursement math applied at stabilized occupancy without grossing up opex

base-year stop leases require opex to be grossed up to the lease-defined occupancy standard before computing the tenant's share, skipping this step misstates the reimbursement owed on both sides
TI/LC reserve absent or under $1.50/RSF/year on a WALT under five years

at that rollover pace, the property will spend real capital re-tenanting regularly, an absent or undersized reserve overstates distributable levered cash flow
Sublease shadow space excluded from effective market rent

tenants marketing space below their own contractual rent are signaling coming vacancy and downward rent pressure that the direct rent roll won't show until it's too late to reprice the deal
Net effective rent more than 15% below face rent

high TI + free rent burns more of stated rent than the pro forma assumes
Rollover concentration above 30% in any single year

single-year rollover exposure in a soft office market is the dominant underwriting risk
Sublease availability exceeding 10% of market stock

tenants are offloading space — direct rents will follow down
Operating expense pass-through base year inconsistency

mismatched base years mean recoveries don't actually protect landlord

Example — office underwriting model walkthrough

Take a 310,000 RSF Class A office tower, 82% occupied with a 4.1-year WALT, priced at $71.3M, $230/RSF. The stacking plan shows 254,200 occupied RSF at an in-place full-service gross rent of $34.50/RSF, producing gross potential rent of $8,769,900. The base year for opex recovery is set at $12.10/RSF; current-year opex, grossed up to the lease-defined 95% occupancy standard, runs $13.35/RSF against 294,500 grossed RSF, or $3,930,575.

Reimbursement income is the delta over the base year, computed only against occupied RSF: (13.35 − 12.10) × 254,200 = $317,750. Adding $279,620 of parking and other income brings effective gross income to $9,367,270. Subtracting the grossed-up opex leaves Year 1 NOI of $5,436,695, a 7.63% going-in cap rate on the $71.3M basis.

Two flags precede the exit assumptions. First, 28,000 SF of sublease space is being marketed at $18/RSF against a $34.50/RSF direct asking rent, a 48% discount and a signal of demand softness the rent roll alone wouldn't show. Second, with a 4.1-year WALT, the model reserves $1.85/RSF/year ($573,500 annually) for TI and leasing commissions at rollover, a separate line from NOI that materially changes levered cash flow available for distribution. The reserve splits new-lease and renewal assumptions, since renewal TI runs materially lighter.

Debt sizing reflects the rollover risk directly: a 1.30x DSCR floor against NOI net of the TI/LC carve-out implies roughly $58.5M of proceeds, but the model caps leverage at 55% LTV, or $39.2M, a deliberately conservative constraint given the WALT. Debt service at those proceeds clears the 1.30x floor with room, which is the point of the lower leverage. At exit in Year 5, re-leasing to 90% occupancy at a market rent of $36/RSF with fresh terms lifts NOI to roughly $6.15M; the exit cap is set 50 bps above entry at 8.13% to price the rollover-heavy underwriting, producing a terminal value near $75.7M. Skipping the gross-up would misstate reimbursement income against the lease-defined 95% standard, and skipping the reserve would overstate distributable cash by $573,500 a year.