Valuing an Office Building: A Lease-Roll Walkthrough

Introduction

Agree on a building's in-place NOI and you can still land twenty percent apart on its value. The gap almost never comes from the cap rate or the discount rate. It comes from the rent roll: the unit-by-tenant table of in-place leases that drives every downstream number, and the assumptions about how those leases roll over (renewal odds, downtime, mark-to-market, concessions) that each analyst has to supply. Valuing an office building is, more than anything, the discipline of turning that lease roll into a defensible view of recurring income and then into a value.

The mechanics run in a clear sequence. Extract in-place income from the rent roll, bridge it to stabilized NOI (the building's recurring earning power under normalized conditions), then triangulate a value across three methods: direct capitalization, a discounted cash flow that models each lease event explicitly, and sales comparison against recent trades. The worked example throughout is a 500,000 SF Class A Manhattan office building with average in-place rent of roughly $80 per SF, which lets the numbers stay concrete from the rent roll to the final value conclusion. This is also the most common technical prompt in office-focused capital markets and advisory interviews, because it forces the candidate to connect lease mechanics, cap rate calibration, and triangulation in one coherent answer.

The starting point is the property's current rent roll: tenant name, suite, square footage, lease commencement, lease expiration, current rent per SF, escalator schedule, lease type (FSG, MG, NNN, etc.), free-rent remaining, renewal options, and other key lease terms. The mechanics of reading that table and bridging it to recurring income are covered in detail in the article on rent roll analysis and the stabilized NOI bridge. A typical institutional rent roll for a 500,000 SF Class A Manhattan office building might show 25-50 individual tenants ranging from a few hundred to 50,000+ SF each, with the largest tenants representing a meaningful share of total income.

A 500,000 SF building with average in-place rent of $80 per SF on 92% occupancy generates gross income of approximately $36.8 million (500K x $80 x 92%). Adding ~10% expense recoveries and other income brings effective gross income (EGI) to roughly $40 million. Operating expenses run $15-20 million (typically $30-40 per SF on Manhattan Class A office), producing in-place NOI of approximately $22-25 million.

In-place NOI is the starting point; stabilized NOI is the analyst's view of the building's recurring earning power under normalized conditions. The bridge typically adjusts for:

Stabilized NOI (Office)

The analyst's projection of an office property's recurring NOI under normalized market conditions, including mark-to-market adjustments on rent rolls, stabilized vacancy assumptions, normalized expenses, and recurring capex reserves. Stabilized NOI is the appropriate input for direct capitalization valuation and is typically 5-15% higher or lower than current in-place NOI depending on whether in-place rents are above or below market and whether the property is currently in lease-up.

Direct capitalization divides stabilized NOI by an appropriate cap rate. Class A office in 2026 traded across roughly a 6.0-8.0% range, with trophy Manhattan core compressing toward the tight end (around 5.5-6.5%) as capital crowded back into the best assets. A property with $22 million of stabilized NOI valued at a 6.0% cap rate trades at approximately $367 million, or roughly $733 per SF.

The cap rate is not a number the analyst can pull from a table; it is calibrated against recent comp sales and the building's own quality, lease structure, and submarket. The 1740 Broadway distressed sale at $308 per SF is one data point, representative of distressed mid-tier office rather than Class A trophy. The 245 Park Avenue recovery at a $2 billion valuation on roughly 1.8 million SF (about $1,100 per SF) sits at the trophy end. Triangulating across several comps, adjusting for submarket, building quality, lease structure, and time, produces a defensible cap rate with typically +/- 50 basis points of uncertainty. The mechanics of how that rate is built up, and what moves it, are covered in more depth in the article on what actually drives cap rates.

The DCF projects cash flows year by year, capturing the specific lease-roll dynamics that direct cap cannot. Each lease expiration requires explicit modeling of:

The DCF discount rate runs 7-9% for unlevered IRR on stabilized core office in trophy markets, higher for value-add or repositioning candidates. The terminal value applies an exit cap rate typically 25-50 basis points wider than the going-in cap to account for asset aging, producing the year-N+1 NOI capitalized at the exit cap.

The sales comparison method uses recent comp sales of similar Manhattan Class A office buildings, adjusted for submarket, building quality, age, lease structure, and time, following the same adjustment discipline laid out in comparable sales analysis. The standard unit of comparison is price per square foot. Three to five comp sales within the prior 12-18 months in the same submarket produce a triangulation range. For a Manhattan Class A trophy building in the example above, a $700-$1,100 per SF range across the comp set is typical, with the specific subject's positioning within the range driven by quality-of-asset adjustments.

The reconciliation step explicitly compares the values from each method:

The reconciled value sits in the $367-400 million range with the income approach and DCF receiving primary weight given the stabilized condition. The recommended value is typically $370-380 million (the midpoint of the income and DCF methods, with sales comp as supporting confirmation). The cost approach at $500M acts as a ceiling check rather than a primary method on stabilized core property.

The lease-roll schedule is the single largest source of valuation variance across analysts looking at the same office building. Two analysts applying the same direct cap method to the same in-place NOI can produce values within 2-3% of each other; the same two analysts running DCF models with different lease-roll assumptions (renewal probability, downtime, mark-to-market rents, TI/LC packages) can produce values that diverge by 15-25%. The asymmetry exists because direct cap assumes stabilized perpetual income while DCF captures the specific lease-roll trajectory.

The implication for analyst work: the lease-roll modeling assumptions are the area where defensibility matters most. A direct-cap-only valuation is typically faster but less informative; a DCF-with-detailed-lease-roll model produces a more accurate valuation but requires defensible assumptions on every major lease event. Sophisticated office analysts run both methods and use the divergence (or convergence) as a diagnostic on whether the lease-roll assumptions are reasonable.

Sensitivity Analysis on Key Inputs

Sophisticated valuation work explicitly tests the sensitivity of the recommended value to changes in the most consequential inputs. The standard sensitivity tests:

The sensitivities together show that the value range from a defensible valuation typically spans a 15-25% spread between the bull case and bear case, even when the base case inputs are well-supported. The recommended value is the base case; the sensitivity range communicates the uncertainty band.

Discount Rate Selection for the DCF

The discount rate selection is one of the most analytically loaded decisions in any DCF valuation. The standard frameworks:

The three approaches typically converge within 50-100 basis points for stabilized core assets. The convergence is part of the discipline; a discount rate that differs materially from the market-derived alternative requires explicit justification. The build-up logic mirrors the discount-rate construction used in corporate-finance DCFs, treated in full in the valuation guide, but property-level work leans harder on the market-derived check because the comparable transactions are usually fresher and more directly observable than equity comparables.

Going-In vs Exit Cap Rate Spread

The differential between the cap rate applied to year-1 stabilized NOI (going-in cap) and the cap rate applied to forward NOI at the assumed exit (exit cap), expressed in basis points. Standard convention assumes exit cap is 25 to 50 basis points wider than going-in cap to account for asset aging over the hold period (the building is older, has shorter remaining lease durations, and may face increased capex needs). Aggressive underwriting sometimes assumes flat or compressing exit caps; lender stress tests typically apply 50-100 bps of exit cap widening as a downside scenario.

The completed valuation feeds into specific transaction decisions across the office market. Acquisition bids: a sponsor's maximum bid price is the value the sponsor's DCF supports at the fund's target IRR, often discounted slightly for negotiation flexibility. The bid is typically 5-10% below the sponsor's full DCF value to leave room for upward negotiation. Disposition asking prices: a REIT or owner pricing an asset for sale typically asks slightly above the seller's DCF value to leave room for downward negotiation. The bid-ask spread is the negotiation zone where most transactions clear.

Loan-to-value sizing: a lender's maximum loan amount is the lower of the LTV-constrained maximum (typically 60-70% of the lender's appraised value, which may differ from the sponsor's DCF value) and the DSCR-constrained maximum (annual NOI divided by target DSCR multiplied by acceptable debt service). The valuation produces the LTV input; the NOI and rate environment produce the DSCR input.

Scaling the Single-Building Valuation to a REIT Portfolio

The single-building exercise becomes a portfolio problem when the asset in question is one of dozens. When an office REIT weighs a take-private offer, the board's financial advisor produces an independent valuation that informs the fairness opinion, and that valuation is essentially a sum-of-the-parts of the individual building values, adjusted for entity-level overhead, debt structure, and other corporate factors. A REIT with 100+ buildings means 100+ valuations, each with its own lease roll and cap rate, rolled up to a portfolio NAV. Specialized teams at investment banks and at independent shops like Green Street Advisors do this work as standard coverage of names such as those covered in the article on the listed office REIT landscape.

At that scale, building-by-building precision collides with practicality. Most REIT-level work applies cap rates by submarket cluster rather than per building, accepting some approximation error for tractability. The judgment call is where to spend the granularity: a contested REIT-on-REIT merger or take-private bid justifies building-level work on the material assets, while a routine equity-research NAV refresh leans on cluster-level aggregation to keep pace. The same underlying methodology serves both; what changes is how deep the analyst goes on any one asset.

The broker side runs the same mechanics from the opposite chair. A broker preparing a building for sale builds the offering memorandum, which lays out stabilized NOI, the broker's cap rate view, the lease-roll schedule with mark-to-market commentary, and a preliminary value range. Brokers tend to present toward the high end of the defensible range to draw out aggressive bids, while sponsor-side underwriters work toward the low end to preserve margin. The clearing price lands somewhere inside that range, shaped by which buyer is most aggressive and which seller is most motivated.

Whichever chair the analyst sits in, the valuation comes back to the same place it started. Direct cap, DCF, and sales comp all run off the rent roll, and the spread between two credible answers traces almost entirely to how each analyst read the lease roll forward. Get the rent roll and its roll-forward right and the three methods converge; get them wrong and no amount of cap rate precision will rescue the number.