Key takeaways
- Going-in cap rate is in-place (or year-1) NOI divided by purchase price. Exit cap rate is the NOI you expect at sale divided by the price you expect to sell for.
- Most underwriters set the exit cap above the going-in cap. A spread of 25 to 75 basis points is a common rule of thumb, and it is a rule of thumb.
- On a $40,000,000 purchase at a 6.0% going-in cap, a 6.5% exit instead of 6.0% takes $3.56 million off the sale price and about 1.5 points off the unlevered IRR.
- Cap rate compression raises values without any change in the building. A buyer who paid for compressed cap rates is exposed when they expand again.
- Check every exit cap assumption against the comps your own firm has seen, including the deals you passed on.
A capitalization rate is net operating income divided by value. It is the unlevered yield a buyer accepts for one year of a property's income, and it moves inversely to price: a lower cap rate means a buyer is paying more for each dollar of NOI.
That is all the general definition you need. The two cap rates that matter in an underwriting model are the one you buy at and the one you assume you sell at.
What the going-in cap rate is
The going-in cap rate is the NOI a property produces at acquisition, divided by the purchase price.
Going-in cap rate = NOI at acquisition / Purchase price
Some firms use in-place NOI (the trailing 12 months, normalized). Others use year-1 NOI from the model. Say which you mean, because on a building with rollover or free rent the two can differ by a full point.
Buy a building for $40,000,000 that produces $2,400,000 of NOI and your going-in cap rate is 6.0%. It is a statement about the price you paid relative to the income you are getting today.
What the exit cap rate is
The exit cap rate, also called the terminal cap rate or reversion cap rate, is the cap rate you assume the next buyer will pay at the end of your hold. Divide the forward NOI (usually the year after your sale year) by that rate and you have your assumed sale price.
Sale price = Forward NOI at sale / Exit cap rate
Hold the same building 5 years with NOI growing to $2,780,000 in the year after sale, and assume a 6.5% exit. Sale price is $2,780,000 / 0.065, or $42,769,000.
The exit cap is an assumption about a market 5 or 10 years away, and no one can know it today. The whole question is how to pick one you can defend.
Why the exit cap is usually set above the going-in
Three reasons, and they stack.
The building is older. A 5-year hold adds 5 years of age to the roof, the systems and the finishes. All else equal, buyers pay a little less for an older asset.
The risk is unknown. You can see today's rates, today's tenant demand and today's buyer pool. Projecting them forward adds uncertainty, and underwriting convention charges for uncertainty by assuming a higher exit yield.
Conservatism is cheap insurance. An exit cap above the going-in cap means the model's returns do not depend on the market getting better. If it does, the upside is yours.
A spread of 25 to 75 basis points over the going-in cap is a common rule of thumb. Treat it as that: a starting point that your comps should confirm or override, and one that varies with sector, hold period and where you think the cycle is.
What the spread does to IRR
Returns in a 5- to 7-year hold are dominated by the sale. A small change in the exit cap moves the sale price by millions and the IRR by whole points, which is why investment committees go straight to that cell.
Stay with the $40,000,000 example: $2,400,000 NOI, 6.0% going-in, NOI growing about 3% a year, $2,780,000 forward NOI at sale after 5 years. Three exits:
Exit cap Sale price Gain on price Unlevered IRR Equity multiple
6.5% $42,769,000 $2,769,000 7.5% 1.39x
6.0%* $46,333,000 $6,333,000 9.0% 1.48x
5.5% $50,545,000. $10,545,000 10.6% 1.58x
*(flat to going-in)
The 50 basis point spread between a 6.0% and a 6.5% exit costs $3,564,000 of sale price and about 1.5 points of unlevered IRR. The same 50 basis points the other way earns $4,212,000 and 1.6 points. The building is identical in all three rows; the only variable is the assumed buyer's appetite.
Note the asymmetry. Each 50 basis points of compression is worth more in dollars than the same 50 basis points of expansion costs, because price is NOI divided by the cap rate and the effect grows as the denominator shrinks. That is also why a model with an aggressive exit cap looks so good.
Cap rate compression and expansion
Cap rate compression is a fall in the cap rates buyers accept across a market or sector. Cap rate expansion is the reverse. Both happen without any change to the buildings involved. For one property, the effect on value is:
Value change = NOI / New cap rate minus NOI / Old cap rate
Three things move them. Interest rates, because a cap rate is a yield and competes with the yield on debt and on bonds. Capital flows, because more money chasing the same assets bids cap rates down. Sector sentiment, because buyers will accept a lower yield on what they expect to grow (industrial for most of the last decade) and demand a higher one on what they don't (secondary-market office since 2020).
For a seller, compression is the best thing that can happen. The NOI you built is capitalized at a lower rate and the sale price rises, as the 5.5% row above shows.
For a buyer, it works the other way around. If you bought at a 5.0% going-in cap because the market had compressed to 5.0%, you paid $48,000,000 for the same $2,400,000 of NOI. If cap rates expand back to 6.0% during your hold, a 6.0% exit on $2,780,000 is $46,333,000, below what you paid, and NOI growth of 16% has produced a loss on price. The buyer who pays for compression is betting it stays compressed.
How to sanity-check an exit cap assumption
The exit cap in a model should be able to answer one question: what did comparable buildings in this market actually trade at, and how does that compare to what the model is assuming for 5 years out?
Pull your own comps first. Every deal your firm has underwritten in the market, including the ones you passed on, came with a price and an NOI. That is a going-in cap rate observed in the field. If your passed deals in this submarket cleared between 5.8% and 6.4% over the last 2 years, a 5.5% exit needs a specific reason.
Compare to the going-in cap on the deal in front of you. An exit below the going-in cap says the market will be better at sale than it is at purchase. Sometimes that is the thesis. Say so in the IC memo rather than letting the cell imply it. The annotated example shows how the exit cap shows up in an IC memo: the spread, the comps behind it and the sensitivity grid.
Test the sensitivity. Run the exit at plus and minus 50 basis points from your base case and show all three. If the deal only works at the compressed exit, that is the finding.
Check the hold period. A 10-year exit cap carries more uncertainty than a 3-year one, and a flat or compressed exit assumption on a long hold needs a stronger case.
The pro forma cap rate trap
Offering memorandums quote a cap rate on the cover. Check which NOI it is on before you repeat it.
If the OM says "6.0% cap rate" and the price is $43,333,000, the NOI behind that number is $2,600,000. If the T12 shows $2,400,000, the broker has capitalized a pro forma NOI that includes rent increases, lease-up of vacant space or expense cuts that have not happened yet. Your actual going-in cap rate on in-place income is $2,400,000 / $43,333,000, or 5.5%.
That is a half point of yield that exists on paper only. Rebuild the cap rate from your own normalized NOI every time. The how to read an offering memorandum post goes through the rest of the cover-page checks, and the how to underwrite a commercial real estate deal guide places the exit cap among the other assumptions an IC will push on.
Where Altrio fits
Every deal a broker sends your firm carries a price, an NOI and therefore a cap rate, whether or not you pursue it. Altrio captures those deals from the team's inbox and keeps them, passed or not, as comps searchable by market. When an exit cap assumption needs checking, the record of what this submarket has actually been offered at over the last 2 years is already in the firm's own data. See market data for how the comps database is organized.
Want your firm's own comps on hand when the exit cap comes up at IC? Altrio keeps every deal you've been sent, by market. Start free
Frequently asked questions
What is a good going-in cap rate?
There is no universal figure. A going-in cap rate is good or bad only relative to recent trades for similar assets in the same submarket, and to the return the buyer needs. Core assets in primary markets trade at lower cap rates than value-add assets in secondary ones.
Should the exit cap rate be higher or lower than the going-in cap rate?
Usually higher, by a small margin, to reflect an older building and an uncertain future market. A spread of 25 to 75 basis points is a common rule of thumb. An exit cap below the going-in cap is a bet on market improvement and should be stated as one in the memo.
What is the difference between exit cap rate and terminal cap rate?
Nothing. Exit cap rate, terminal cap rate and reversion cap rate all mean the cap rate applied to forward NOI to estimate the sale price at the end of the hold. Different firms and different textbooks use different names.
What is the difference between cap rate compression and cap rate expansion?
Compression is a fall in market cap rates, which raises values for the same NOI. Expansion is a rise, which lowers them. Using the example above, 50 basis points of compression on $2,780,000 of NOI adds $4,212,000 of value, and 50 basis points of expansion takes away $3,564,000.
What causes cap rate compression?
Falling interest rates, more capital allocated to real estate, and improving sentiment toward a sector or market. Each one leads buyers to accept a lower yield for the same income, which pushes prices up. Expansion is caused by the reverse.
How does cap rate compression affect IRR?
Compression during the hold raises the sale price and the IRR; expansion lowers both. In the example above, a 50 basis point move in the exit cap changes the unlevered IRR by roughly 1.5 points on a 5-year hold. Shorter holds are more sensitive to the exit cap than longer ones, because the sale is a larger share of total cash flow.
Which NOI do you use for the exit cap rate?
Most models use the NOI from the year after the sale year, since a buyer prices on the income they will receive. Some use the sale-year NOI. Pick one, apply it consistently across deals, and state it in the assumptions page.
