Most brokers' offering memorandums read like someone trying to sell me something, because they are. I have learned to skip the rent-growth story and go straight to the last page of assumptions, where the exit cap rate lives. That one number tells me more about how a deal was underwritten than any three pages of rent comps.
The exit cap rate — also called the terminal cap rate — is the cap rate the model assumes the buyer of your buyer will pay when you sell. It sets the assumed sale price: stabilized NOI divided by the exit cap. A small move in that denominator swings the sale price by a lot, and the sale price is usually the biggest single piece of the projected return. So the question is never just 'what's the entry cap.' It's 'what did you assume about the cap five or seven years from now.'
The standard practice, and the honest one, is to underwrite the exit cap above the entry cap. The common rule of thumb in the industry is to add roughly 10 basis points of expansion for each year of the hold period — so a property bought at a 5% cap on a five-year hold might be sold in the model at a 5.5% exit cap. The reason is simple: the building will be older, the market is uncertain, and a conservative exit protects the return from depending on cap rate compression that may never arrive.
Recent investor surveys tell me I'm not alone in this. A 2026 survey of multifamily investors found that nearly half are underwriting exit cap expansion of 25 to 50 basis points over their going-in cap rates, and about three-quarters are capping rent growth assumptions at 2.5% over the next 18 months. That's a market saying the same thing I do: the money has to come from the operations, not from hoping the market bails you out.
What I actually watch for in an OM is the reverse: an exit cap below the entry cap with no real explanation. That means the seller is assuming cap rate compression — that the market will pay a lower yield for the same building in five years than it does today — and baking that assumption into the projected IRR. Sometimes there's a real thesis behind it: a genuine repositioning into a better asset class, a market with contracting supply. Most of the time there isn't, and it's the single most common way a model manufactures a return that doesn't exist.
The discipline matters because of how the math compounds. Take a building bought for $10 million with $500,000 of NOI, a 5% entry cap. Grow NOI to $600,000 over five years. Sell it at a 6% exit cap and the sale price is $10 million — all the value came from the income you grew, and the model admits it. Sell it at a 5% exit cap and it's $12 million, and $2 million of that 'value creation' is just an assumption about what strangers will pay in the future. One of those two models will teach you something when it breaks. The other will just break.
So when I'm looking at a deal, I ask for three things before anything else: the entry cap, the exit cap, and the spread between them with the reasoning written down in one sentence. If the spread is negative and the sentence is vague, I move on. The business plan — the turns, the collections, the expense work — is what I'm buying. The exit cap is just where I check whether the model believes that too.
← All insights