Cap rates and the cost of waiting

Thirty-two billion dollars changed hands across roughly 1,450 U.S. apartment properties in the first quarter of 2026, at an average cap rate of 5.75 percent and $204,061 a unit, according to RealPage's read of the RCA data. Quarterly volume was 42 percent below the fourth quarter of 2025 and well under the five-year quarterly average of $55 billion. Look at the trailing four quarters instead of the noisy single one and the picture flips: $170.4 billion changed hands over the year ending in the first quarter, up 6 percent, across 7,256 properties, up 14 percent.

Two true statements about the same market, pointing in different directions depending on the window you pick. That is the whole cap-rate story right now, and it is why "the cost of waiting" is not a rhetorical question.

The survey said compression. The first quarter did not deliver it.

CBRE's own year-ahead call, published in February, was for cap rates to compress 5 to 15 basis points across most property types in 2026, with better assets seeing more of the move. Three months into the year the market gave a different answer. MSCI's data on the first quarter found deal volume climbing at double-digit rates across most sectors, which was the good news the CBRE forecast implied, but cap rates drifted higher across most sectors over the same period, sitting in tension with that healthier volume. Higher volume and higher cap rates at the same time is not the combination the February forecast was underwriting.

Neither reading is wrong. They are measuring different things. The survey captures sentiment, professionals estimating where pricing sits and where they expect it to go, drawn from 3,600 cap rate estimates across more than 50 U.S. markets. The transaction data captures what actually closed. When those two diverge, the honest conclusion is that nobody, including the professionals who built that survey, actually knows which way this resolves yet. CBRE's own H2 2025 write-up already flagged the tell: a historically wide spread persists between the 25th and 75th percentile of estimates, with only early signs that the top end is starting to come down. A wide spread between what optimists and pessimists think an asset is worth is a market that has not actually agreed on a price. It just agrees that averages exist.

The bid-ask spread narrowed. That part is real and it happened anyway.

Here is the genuinely interesting part, because it happened despite the cap-rate confusion, not because of it. JLL's Global Bid Intensity Index found investor bidding competitiveness across multifamily, industrial, retail, and office had narrowed to its smallest spread in more than three years by March 2026. Buyers and sellers converged on price even while the underlying cap rate direction stayed genuinely unresolved. That convergence is what is actually driving the trailing-year volume growth, not a definitive cap-rate call in either direction.

It also happened while the risk-free rate moved against the "cap rates fall" thesis, not with it. The 10-year Treasury sat near 4.79 percent in January 2025, fell to roughly 4.24 percent by midyear 2025, the level CBRE's H1 2025 survey was built against when it recorded an all-property average cap rate of 6.84 percent, down 9 basis points from the prior survey. By July 2, 2026, the 10-year was back up to 4.49 percent, a 25-basis-point round trip against the trough the compression forecast was implicitly anchored to. Add the Fed's own June dot plot, raised rather than cut at its most recent meeting, and the tailwind the February cap-rate forecast assumed is smaller today than it was when the forecast was written.

So the bid-ask spread closed on vibes and price discovery, not on a friendlier discount rate. That is a fragile kind of healing. It can survive a flat rate environment. It is not obviously built to survive a Committee that just told you it might hike.

CBRE's forecast leaned optimistic for a real reason, not a manufactured one, and the underlying sentiment behind it has not gone away even though the print has not caught up to it yet. The same 2026 outlook that called for compression also projected total U.S. investment volume rising 16 percent to $562 billion, nearly back to the 2015-2019 pre-pandemic annual average, with 74 percent of surveyed investors planning to buy more this year than last. Capital genuinely wants back into this market. The disagreement is only about whether that appetite shows up first as tighter cap rates or first as more volume at whatever cap rate the market actually clears, and the first quarter's data says volume is arriving before the compression, not alongside it.

That ordering has a mechanical explanation as much as a psychological one. Capital that has been sitting out for two years does not wait for the perfect entry cap rate before it acts; it acts once the bid-ask gap narrows enough to close a deal at all, and worries about the exact basis point later. Compression, if it comes, tends to show up after volume has already normalized, once enough comparable trades exist for appraisers and lenders to underwrite to a tighter number with confidence. Reading Q1's wider cap rate as bearish, on its own, mistakes the leading indicator for the lagging one.

What does this actually do to when ground breaks?

Three groups make decisions off this number, and each one is exposed differently.

A seller holding a stabilized asset is deciding whether this quarter's bid-ask convergence is durable enough to sell into, or whether waiting for the CBRE-forecast compression captures another 5 to 15 basis points of value. Wait too long and the compression that was supposed to show up in "most of 2026" instead shows up as the higher cap rate MSCI already measured in the first quarter, and the seller has traded a live, narrowed bid-ask window for a hypothetical one that may not arrive.

A buyer underwriting a new acquisition is deciding whether today's 5.75 percent multifamily print is a floor or a moment. Underwrite to further compression that does not show up and the deal is priced for an exit that is not there. Underwrite to the first quarter's realized direction, cap rates staying flat to slightly wider, and a deal that only works at CBRE's forecast compression should not get done at all.

A developer deciding when to break ground is exposed to both sides at once, which is the case we actually care about here. We have already quantified what a slow entitlement timeline costs in raw carry at today's rates. Layer a cap-rate call on top of that carry cost and the arithmetic gets harder, not easier: a project that pencils at a 5.75 percent exit cap and today's construction-loan spread can go negative on a 25-to-50-basis-point adverse move in either the cap rate or the long rate, and 2026 has already shown both can move against a developer inside a single quarter.

Waiting is not free. It never has been. What is new is that waiting no longer has an obvious direction to wait toward. For most of the last cycle, "wait for rates to come down" was at least a coherent bet, even when it was a bad one. Today the professionals who write the cap-rate surveys and the professionals who booked the first-quarter trades are telling you two different things, and the Fed just told you its own forecast moved against the friendlier one.

My own read: the CBRE forecast is more right than the first quarter's print, but for a boring reason rather than a hopeful one. Q1 volume is thin and seasonally distorted enough that a handful of large, wide-cap-rate deals can move the quarterly average without moving the market. I expect the trailing-four-quarter data to keep showing the compression CBRE called for by year end, even though any single quarter along the way, including a possible second-quarter print, may say otherwise. If the full-year number comes in flat or wider instead, I was wrong, and the honest read for 2027 is that this cycle's cap-rate story never actually turned, it just found new professionals willing to say it had.

Do not build a pro forma that needs the compression forecast to be right. Build one that survives either answer, because the market currently cannot tell you which one it is going to give you, and the two most-cited sources on this exact question disagree with each other in the same quarter.

This analysis is a source-cited research summary drawn from public records, not legal advice. It can contain errors and should be verified independently before any investment decision.