Cap Rate Compression Is Over: Cap Rate Underwriting in a Higher-Rate World

For a decade, cap rate underwriting had a silent partner: falling yields. You could underwrite a mediocre going-in spread and still hit your return because the exit cap came in tighter than the entry. That partner has left. With the 10-year Treasury sitting near 4.25 to 4.50 percent and CBRE reporting average cap rates that barely clear borrowing costs, the model that carried the last cycle now produces losses. In a higher-rate world, cap rate underwriting has to earn its return from operations, not from a repeat of compression that is unlikely to come.

The thesis is simple and uncomfortable. Multiple expansion was doing work in your model that you attributed to skill. Strip it out, and the deals that still pencil are the ones where net operating income growth and the going-in spread carry the whole return.

Key Takeaways

Cap rate compression is no longer a reliable underwriting assumption. CBRE's H2 2025 Cap Rate Survey puts multifamily near 5.3 percent and office near 6.4 percent, against a 10-year Treasury around 4.25 to 4.50 percent, a spread far tighter than the 2010 to 2020 historical average.
When the average CRE cap rate barely exceeds the average borrowing cost, positive leverage is thin or negative, and the exit assumption stops subsidizing weak going-in economics.
The disciplined move is to underwrite the exit cap rate at or above the going-in cap rate, then let NOI growth and the entry spread carry the return.
A 50-basis-point move in the exit cap can swing residual value by 8 to 10 percent on a stabilized asset. Small assumptions now dominate the outcome.
Roughly 875 billion dollars of commercial and multifamily debt matures in 2026 per the Mortgage Bankers Association, repricing from older sub-5 percent coupons into today's higher rates.

What changed in cap rate underwriting when rates rose?

Cap rate underwriting changed because the exit stopped being a tailwind. For a decade, exit caps were routinely underwritten below going-in caps, and compression added return the sponsor never operationally earned. With rates higher and spreads tight, that free return disappeared, and models that assumed it now overstate value.

The mechanics are worth stating plainly. A cap rate is a yield: NOI divided by value. When cap rates fall, the same NOI is worth more, so value rises with no operational improvement. Between 2010 and 2021, that repricing was close to continuous, and it flattered every model that assumed the exit would clear tighter than the entry.

CBRE, whose Cap Rate Survey has run for 17 years and draws on roughly 3,600 estimates across more than 50 markets, reported current cap rates in H2 2025 of about 5.2 percent for industrial, 5.3 percent for multifamily, and 6.4 percent for office and retail. Set those against a 10-year Treasury near 4.25 to 4.50 percent and the spread is thin by historical standards. CBRE's own note on estimation points out that the average cap-rate-over-Treasury spread from 2010 to 2020 was roughly 230 basis points for multifamily and 340 for industrial. Today's spreads sit well inside those ranges.

For the underwriter, the conclusion is direct. If the spread that compensates you for risk is already compressed, betting on further compression is betting against the historical mean. See the exit cap rate and going-in cap rate glossary entries for how these two assumptions interact.

Why does the cap-rate-to-borrowing-cost spread matter more than the cap rate itself?

The spread between cap rate and borrowing cost determines whether leverage helps or hurts. When the going-in cap rate exceeds the loan rate, leverage amplifies equity yield, which is positive leverage. When borrowing costs approach or exceed the cap rate, leverage drags returns, and the deal must earn its way out through NOI growth rather than financial engineering.

Market data has narrowed this gap to almost nothing. Reporting through 2026 puts average CRE borrowing costs near 6.24 to 6.57 percent against average cap rates around 6.34 percent, an unusually tight relationship. When the going-in yield and the cost of debt sit within a few basis points of each other, day-one positive leverage is marginal or gone.

That is the number to stress. Consider a simple worked example on a stabilized asset:

Assumption	Compression-era view	Higher-rate discipline
Purchase price	10,000,000	10,000,000
Year 1 NOI	550,000	550,000
Going-in cap rate	5.50%	5.50%
Exit cap rate assumed	5.00%	5.75%
NOI at exit (5-yr, 3%/yr growth)	637,600	637,600
Residual value at exit	12,752,000	11,088,700
Value from NOI growth vs compression	Compression adds ~1,663,300	Compression adds nothing

The compression-era column books more than 1.6 million dollars of residual value that comes entirely from a tighter exit cap, not from anything the operator did. Arithmetic check: 637,600 divided by 0.05 equals 12,752,000; 637,600 divided by 0.0575 equals 11,088,696, rounded to 11,088,700. The difference, 1,663,304, is the compression premium. Underwrite it out, and the deal has to justify itself on operations.

How should you underwrite the exit cap rate now?

Underwrite the exit cap rate at or above your going-in cap rate as the base case, and widen it further for longer holds, transitional risk, or secondary markets. This removes compression from the return and forces the model to earn its equity yield from NOI growth and the entry spread, which are the only levers you actually control.

A workable rule set for cap rate underwriting today:

Base case exit cap equals going-in cap plus 0 to 25 basis points. Compression is a bonus you do not underwrite.
Add 10 basis points of exit cap for each year of hold beyond five, reflecting depreciation and re-tenanting risk on an aging asset.
Stress the exit cap 50 basis points wider than base and confirm the deal still clears its return hurdle. If it only works at the tight exit, it does not work.
Reconcile the going-in cap to actual, verified NOI, not to a pro forma. An inflated NOI understates the true entry cap and hides the problem.

As one CBRE Cap Rate Survey summary framed the turn, the survey now suggests a new market cycle on the horizon rather than a continuation of the last one. For the underwriter, a new cycle means the last cycle's assumptions expire with it.

Here is the expert-voice line worth keeping on the whiteboard: in a higher-rate world, the exit cap is not where you find upside, it is where you protect yourself from being wrong.

What does the 2026 maturity wall mean for exit assumptions?

The 2026 maturity wall means many assets will change hands or refinance into today's rates, resetting the comparable transactions that inform exit caps. The Mortgage Bankers Association reports roughly 875 billion dollars of commercial and multifamily debt maturing in 2026, near 17 percent of outstanding loans, much of it repricing from sub-5 percent coupons.

Reporting through 2026 describes older debt averaging near 4.76 percent maturing into new originations around 6.24 percent, a rate shock of 150 basis points or more. Some estimates put total 2026 maturities well above 1.5 trillion dollars. That volume matters to your exit assumption because forced and semi-forced sales set the comparable set the next buyer underwrites against.

If a wave of owners must transact into higher financing costs, buyers will demand wider going-in caps to make deals pencil, which is the mechanical opposite of compression. Underwriting an exit cap tighter than your entry in that environment assumes the maturity wall resolves without repricing values. The historical base rate for that assumption is poor. The underwriting model should treat the wall as a reason to widen the exit, not to hold it flat.

Frequently Asked Questions

Is cap rate compression really over, or just paused?

Compression as a dependable underwriting assumption is over for now. Cap rates could tighten if rates fall, but CBRE's forecast for most property types in 2026 is modest movement of 5 to 15 basis points, not the sustained compression of the 2010s. Underwrite the base case without it and treat any tightening as upside.

What exit cap rate should I use if I do not know where rates are going?

Set the exit cap equal to or slightly above your verified going-in cap rate, then stress it 50 basis points wider. If the deal clears its return hurdle at the wider exit, the assumption is defensible. If it only works at a tighter exit, the return is a bet on compression, not on the asset.

How much does the exit cap rate move residual value?

On a stabilized asset, a 50-basis-point change in the exit cap typically swings residual value by 8 to 10 percent. In the worked example above, moving from a 5.00 percent to a 5.75 percent exit cap cut residual value by roughly 1.66 million dollars on a 10 million dollar purchase, more than 13 percent.

Conclusion

Cap rate compression flattered a decade of underwriting and let mediocre deals look like skill. That subsidy is gone. With cap rates barely clearing borrowing costs and roughly 875 billion dollars of debt repricing in 2026, the exit is a source of risk, not return. The discipline is to underwrite the exit cap at or above the entry, stress it wider, and require the deal to earn its return from NOI growth and the going-in spread. Deals that survive that test were always the good ones. The higher-rate world simply stops paying you to pretend the others were too.