Cap rates and net operating income, explained properly

Net operating income, built from the rent roll

Every real estate valuation starts at the property, and the property starts with net operating income. NOI is the cash a building produces in a year from operating it as a building, before anything to do with how it was financed, how it is being improved, or who owns it. You build it from the rent roll, the lease by lease schedule of who occupies what space, what they pay, and when they roll.

The build is mechanical, and an interviewer who asks for it wants the lines in order:

The recovery line is where candidates first slip. Under a triple net lease the tenant pays taxes, insurance, and maintenance directly, so the recovery and expense lines are both small; under a gross lease the landlord pays and bills back, so both are large. NOI can be identical either way, which is why you ask what lease structure sits under the numbers before comparing assets.

What matters more is what NOI is deliberately before. NOI sits above debt service, capital expenditure, tenant improvement allowances and leasing commissions, corporate general and administrative expense, depreciation, and income taxes. Every exclusion is intentional. Stripping out debt service makes NOI comparable across owners with different capital structures. Stripping out corporate overhead makes it a property number rather than a company number. Stripping out depreciation keeps a non cash charge from distorting an asset whose value is driven by income, which is the same reason net income is a poor earnings measure for a REIT.

The exclusion that causes real trouble is capital expenditure. Buildings consume capital just to stay where they are. Roofs, elevators, HVAC systems, and the tenant improvement package and broker commission you pay every time a lease rolls are not optional, and none of them appear in NOI. A property with $10M of NOI that needs $2M a year of recurring capital to hold occupancy is a materially worse asset than one needing $200K, and NOI reports both identically. That gap is why AFFO exists on top of FFO, subtracting recurring capital and straight line rent to get closer to distributable cash. Say that link out loud, that NOI ignores capital spending and AFFO puts it back, and you have connected the property math to the equity math in one sentence. The full version lives in FFO, AFFO, and REIT earnings.

The cap rate equation runs in both directions

The capitalization rate is NOI divided by value. That is the whole definition. Written that way it is a yield: the unlevered, first year return an all cash buyer earns on the purchase price, before financing and before appreciation.

Rearrange it and it becomes a pricing tool, which is how it is used most of the time:

Value = NOI / cap rate

Take the $10M of NOI from the table. At a 5.0% cap rate the property is worth $200M. At a 6.0% cap rate the same $10M is worth roughly $167M. One hundred basis points, a number that sounds small when you say it, just erased $33M of value, about seventeen percent. At a 4.0% cap rate that same income stream is worth $250M.

That convexity is the thing to internalize. Because value is NOI divided by a small number, sensitivity to that number is violent, and it worsens as cap rates go lower: moving from 6% to 5% adds more value than moving from 8% to 7%. It is why sensitivity tables are run as a grid of NOI against cap rate, and why an underwriter arguing about twenty five basis points is not being pedantic.

The arithmetic runs backward too. A building trading for $200M on $10M of NOI just gave you an observed 5.0% cap rate, which becomes a comparable for the next asset you price. Those comparables are the raw input into a net asset value build, the primary methodology in the sector, covered in how real estate companies are valued.

Cap rate versus multiple

A cap rate is close to the inverse of a multiple of NOI. A 5.0% cap rate is a 20.0x NOI multiple. A 6.25% cap rate is a 16.0x multiple. An 8.0% cap rate is a 12.5x multiple.

Keep that translation ready, because you will get generalist interviewers in a real estate group and real estate questions in a generalist interview. To someone who lives in EV/EBITDA, "it traded at a 4.5% cap" lands nowhere and "roughly 22 times property level operating profit" lands immediately. The mapping is imperfect, since NOI carries no corporate overhead and cap rates apply to single assets rather than whole companies, but as a translation device it works.

Why a low cap rate is not the same as an expensive building

Here is the intuition candidates are actually being tested on, and most of them miss it.

A cap rate behaves like a discount rate minus a growth rate. It comes out of the same perpetuity logic as a terminal value: capitalize a growing income stream and value equals next year's income divided by (required return minus growth). Rearranged, the cap rate is roughly the required return on the asset less the expected growth rate of its NOI.

Two consequences fall out of that.

First, a low cap rate can mean the market expects rent growth, not that a buyer overpaid. An asset in a supply constrained market with in place rents below market, where leases roll into higher rents, deserves a lower going-in yield, because the yield you buy today is not the yield you will own in three years. Paying a 4.5% cap is rational if that NOI compounds at five percent while the alternative at a 7.0% cap has flat income forever. Run both to a five year IRR and the low cap rate asset can win.

Second, a low cap rate can mean lower perceived risk rather than growth. Better location, stronger tenant credit, longer weighted average lease term, a more diversified rent roll, less capital intensity, a deeper buyer pool at exit. All of those lower the required return, which lowers the cap rate at the same growth expectation.

So when an interviewer says "this building trades at a 4% cap and that one at an 8% cap, which is cheap," the answer is neither, until you know what is inside them. The 8% asset might be a single tenant building in a secondary market with four years of term left and a tenant nobody wants to underwrite. That 8% is not a bargain, it is compensation. The same logic separates property types across the sub-sector and coverage map, and saying it out loud separates candidates who memorized the formula from candidates who understand it.

Going-in, exit, and what your exit cap assumption signals

Two cap rates show up in any hold period model.

The going-in cap rate is year one NOI divided by the purchase price. It prices what you are buying.

The exit cap rate, also called the terminal or residual cap rate, is the rate you assume a future buyer applies to NOI in the year after you sell. Because it capitalizes a perpetuity at the end of the hold, it drives more of the total return than anything else in the model, often most of an unlevered IRR on a stabilized asset.

Convention is to underwrite an exit cap rate at or above the going-in cap rate, typically twenty five to fifty basis points wider on a five to ten year hold. The building is older at exit, its systems are closer to replacement, and you do not want returns to depend on a market friendlier than the one you bought into. Assuming cap rate compression is assuming multiple expansion, and every credible investment committee treats that as something you defend, not something you get for free.

So a model showing an exit cap below the going-in cap is saying something. Either the sponsor believes the asset gets repositioned into a better risk category (a value-add plan lifting a B asset to A quality, or a lease-up that stabilizes a transitional building), or the returns did not work and someone reached for the easiest lever. Shown that model in an interview, the answer they want is: this deal relies on cap rate compression, so what changes about the asset to justify it, and what is the return at a flat exit cap.

Nominal versus economic cap rate

The headline number can lie, and knowing why is a genuine differentiator at superday.

A nominal cap rate is the number in the press release: stated NOI divided by price. An economic cap rate adjusts that NOI toward what an owner actually receives on a normalized basis. Three adjustments do most of the work.

Free rent and concessions. A lease signed with twelve months free on a ten year term shows full contractual rent in the rent roll while the tenant pays nothing for a year. Straight line accounting spreads the concession across the term, cash does not, so "is that a cash cap rate or a GAAP cap rate" is a good question to ask.

Tenant improvement allowances and leasing commissions. Real dollars spent to produce the rent in the rent roll, and they sit below NOI. An asset with heavy near term rollover carries a large unfunded capital obligation the nominal cap rate does not show.

Below market or above market in place rents. If leases were signed years ago at rents under what the space would fetch today, in place NOI understates the building's earning power, and a buyer will pay a lower nominal cap rate for the right to mark rents up on rollover. The reverse is worse: rents above market look like an attractive yield right up until the leases roll down.

The habit to build is to state which NOI you are dividing. Trailing twelve month, in place, forward twelve month, and stabilized NOI produce four different cap rates on the same building.

What drives spreads between property types and markets

Two structural drivers explain most of the difference in where cap rates sit, and you can discuss both without quoting a market level.

Lease duration and capital intensity. Hotels re-price every night, making lodging income the most volatile and most operationally intensive in the sector, on top of heavy renovation capital. Net lease assets with fifteen to twenty year terms and contractual escalators sit at the other end: bond like income, minimal landlord capital, and a cap rate that behaves more like a credit spread than a property yield. Apartments re-lease annually, repricing fast in both directions. Office and industrial sit between, with multi year leases and meaningful re-leasing capital.

Tenant credit and rent growth expectations. An identical building leased to an investment grade tenant and to an unrated one prices differently, because the income is differently likely to arrive. Supply constraints work the same way: a market where new construction is hard to permit supports a higher expected growth rate, which lowers the cap rate through the discount minus growth relationship.

Financing conditions sit underneath all of it. Cap rates are a yield, and yields compete with other yields, so when debt costs more the levered buyer's bid falls and cap rates face upward pressure. The mechanics of that debt, from mortgages to unsecured REIT bonds to securitized loans, are covered in real estate debt and CMBS basics.

Four ways candidates get marked down on this

Quoting a cap rate off the wrong NOI. Dividing by in place NOI when free rent is burning off, or by trailing NOI on an asset still leasing up, produces a number that is technically arithmetic and practically meaningless. Say which NOI you used.

Confusing a cap rate with a levered IRR. A cap rate is an unlevered, single year, static yield. An IRR is a levered, multi year, cash flow weighted return that depends on financing, growth, capital spending, and the exit. Treating "a 6% cap" as "a 6% return" is one of the fastest ways to lose an interviewer.

Forgetting a capital reserve. Comparing a cap rate to any return that reflects capital spending means subtracting a reserve from NOI first, or you are comparing a number that ignores roofs and tenant improvements to one that does not.

Averaging cap rates across a portfolio. A portfolio cap rate is total NOI divided by total value, an NOI weighted average, not the simple average of the asset level cap rates. Take one asset with $1M of NOI at a 4% cap ($25M of value) and another with $9M of NOI at an 8% cap ($112.5M). The simple average is 6%. The real portfolio cap rate is $10M over $137.5M, about 7.3%. Size, not asset count, drives the number.

Practice question

What is a cap rate, and why would you pay a lower cap rate for one building than another?

A cap rate is net operating income divided by value, so it is the unlevered first year yield an all cash buyer earns on the purchase price. Flip it around and value equals NOI divided by the cap rate, which is how most stabilized assets actually get priced. A property with $10M of NOI at a 5% cap rate is worth $200M, and at a 6% cap rate it is worth about $167M, so a hundred basis points moves value by roughly seventeen percent.

The reason I would pay a lower cap rate for one building than another is that a cap rate behaves like a discount rate minus a growth rate. Buying at a low cap rate means accepting a lower year one yield because I expect the income to grow, or because the risk around it is lower: better location, stronger tenant credit, longer lease term, less capital needed to keep the space leased. A high cap rate is usually not a bargain, it is compensation for flatter or riskier income.

Two things I would keep straight. I would quote the cap rate off stabilized forward NOI, not off in place rent if free rent is burning off or leases sit below market. And a cap rate is not a levered IRR, because it says nothing about financing, growth, capital spending, or the exit.

What the interviewer is listening for: They want the formula stated cleanly in both directions and used on a number, because a candidate who can only recite NOI over value has memorized rather than understood. The real signal is the discount rate minus growth intuition, which separates someone who thinks a low cap rate means expensive from someone who knows it usually means growth or safety. Naming the forward NOI and levered IRR traps unprompted tells them you have seen an actual underwriting model.