After years of false starts and regulatory confusion, a few serious projects are pointing toward what this technology can actually deliver for real estate
In the months that followed the COVID pandemic, institutional real estate investors made a weighty decision: the broad reallocation of capital from office to multifamily that had been underway for more than a decade hadn’t gone fast enough. Building a resilient portfolio for the 2020s and beyond required more urgent action: office assets needed to be jettisoned in favor of apartments, particularly apartment assets in markets with strong fundamentals of job growth and friendly business climates. In other words, Sun Belt multifamily.
Office, in particular, suffered at the hands of capital flows as institutions shifted away from an asset class that many considered risky and out of favor in an increasingly virtual world.
But as 2026 dawns, there are reasons to believe this was a historic blunder. A glut of new supply has pushed rents down nationally year-over-year, with many Sun Belt markets suffering the greatest blows. And the long-term picture is little better, with declining birth rates and anti-immigration sentiment leading to a shrinking pool of humans to fill all those new buildings. At the same time, the office appears poised for a rebound, with a majority of markets now seeing positive net absorption.
So did institutions woefully mistime the market?
How We Ended Up Here
Institutional reallocation away from office and toward multifamily did not begin in March 2020. It was already underway in the early 2000s, after the dot-com bust exposed how cyclical office cash flows could be relative to housing demand, and as “core” portfolios broadened beyond the old office-centric mix. NCREIF’s ODCE index, which measures core equity returns, pegged office’s share of institutional core portfolios at 43% in 2000, declining to 34% by 2020. Multifamily, on the other hand, rose from 18.7% in 2005 to 27% at the end of last year. (Industrial had an even more dramatic rise during that period.)
While office’s volatility through the 2001 and 2008-9 downturns contributed to this shift, the emergence of multifamily as an investable asset class was just as much a part of the story. Prior to the 1990s, institutional investment in multifamily was rare; rental housing made up less than 5% of institutional real estate portfolios in 1988. Apartments were generally owned by local private investors; institutions favored trophy office assets.
But as multifamily emerged as an institutional-grade category through the 1990s, large investors couldn’t get enough, eventually pushing even Class B multifamily cap rates down below those of Class A urban offices — a situation that would’ve been unthinkable to a real estate investor in 1990. To institutions, however, a portfolio of creditworthy, middle-class tenants proved more resilient than a shiny building supported by a handful of corporate leases. And with GSE support, lenders concurred.
By March of 2020, hundreds of billions of dollars in institutional capital had flowed out of office and into multifamily.
But in the months following the pandemic, institutional investors broke the glass and pulled the alarms. Already out of favor, office became a total pariah — an asset class seemingly destined for the real estate dustbin. An office building in the immediate post-pandemic years was viewed as a Rust Belt steel factory in the postwar years, a dinosaur just waiting for the meteor.
Permanent remote work was seen by many as the “new normal,” a technological inevitability unlocked by broadband and video technology, an advance just waiting for some external shock to knock corporate America out of the in-office norm. It did not help that office industry boosters repeatedly pronounced “return to normal” dates — Labor Day 2020, January 2021, Labor Day 2021, January 2022 — that always seemed to fall flat.
Investors also circulated memos and articles detailing office’s historically poor performance over the past decade as justification for fleeing the category. (Of course, the reasoning was somewhat circular: the ongoing institutional reallocation depressed office prices, driving returns lower and giving institutions another reason to avoid the category.)
The end result was predictable: institutions could not get out of the office sector fast enough. ODCE’s office share plunged to 17% by 2025. Many institutions sold office assets at steep losses, willing to take the pain now before things inevitably got worse.
Along with industrial, multifamily was a major beneficiary of this epic reallocation. Billions in institutional capital flowed into multifamily, pushing cap rates down to new lows in 2021 and triggering a wave of development. Flush with institutional capital, developers drove multifamily completions to a peak not seen since the early 1970s.
In many investors’ minds, multifamily wasn’t simply replacing the office as a line in a portfolio allocation; remote work meant that multifamily was literally replacing the office as the place work would get done. New multifamily buildings built thousands of square feet of coworking and rentable conference room space while office buildings couldn’t find buyers.
Where We Are Today
As 2026 dawns, the vibes around multifamily and office appear to be headed in opposite directions — and not the way most institutional investors were hoping.
After five dark years, the office market appears to be turning a corner. It’s worth noting average office rents actually rose: trophy Class A properties continued to lease while many lower-quality office owners refused to capitulate on rents, accepting vacancy — and repeated loan extensions — rather than admit their buildings were worth less money. So while market vacancy rose, lease rates inched upward.
But there is good reason to believe that vacancy is at or near its peak. While 2025’s U.S. net office absorption was still negative to the order of 15 million square feet, that’s a substantial recovery from 2024’s net loss of more than 49 million square feet. And many top individual markets including NYC, San Francisco, and Dallas are seeing positive net absorption. Existing assets will also be buoyed in the coming years by a very dry office construction pipeline.
Sun Belt multifamily is facing a very different situation. The tremendous amount of new supply delivered over the past 24 months in many markets caused rents to fall as multifamily operators were forced to compete for residents. Owners offered lengthy free-rent periods, and lease-ups dragged on as tenants shopped for deals. There is now plenty of rental inventory in core Austin under $1,000 per month, an unthinkable scenario three years ago.
To be clear, multifamily in many other places — the Midwest and coastal markets, for instance — is doing fine. But institutional investors generally avoided those markets, uninspired by the midwest’s lackadaisical growth and spooked by coastal markets’ regulatory outlook. They sought out the Sun Belt, betting on demand and job growth to win the day.
But demand is offering its own challenges.
The Long View
Institutional investors pride themselves on taking the long view. Core investments, by definition, shouldn’t be differentiated by operational savvy or value add; they’re about picking great assets that will stand the test of time. And this was, for many investors, the key justification for reallocating to multifamily. People always need somewhere to live.
But that justification is supported by a key assumption: there will always be more people. And that assumption is on increasingly shaky ground.
Starting around 2008, U.S. birth rates began falling dramatically, dropping from near-replacement rate to well below that by 2020. The impact on real estate thus far has been negligible for two reasons: one, a drop in birthrates today won’t show up in household formation — the real driver of multifamily demand — for at least twenty years. Two, the drop in birth rates was more than counterbalanced by a significant increase in immigration, driving overall population growth up.
But both of those mitigants are disappearing. Anti-immigration sentiment is high, and by all early measures it appears that 2025 will show a remarkable drop in the nation’s immigrant population due to administration policies.
In the coming years the birth rate drop that began in 2008 will start to impact multifamily demand. 2025 was “peak 18 year old.” But subsequent cohorts of young people are substantially smaller than prior ones — a confluence of lower birth rates and the end of the last echo of the baby boom. We wrote about this problem earlier this year but came up with few good solutions or silver linings for the real estate industry.
So while everyone in multifamily is concerned about oversupply, the long-term demand story is the far bigger problem. There simply won’t be enough humans to rent all those units.
Still, the office sector as a whole does not appear to be on its deathbed as many predicted. The notion that the office would be wholly replaced by some combination of remote work and third places is simply not coming to pass.
And unlike multifamily, offices don't have the same long-term headwinds. Office demand is far less coupled to raw population statistics; if anything, one could argue that employers forced to compete for a dwindling pool of workers would encourage an office amenities arms race. It’s likely we’re already seeing this in certain industries as occupiers push office owners to deliver quality in trophy assets in ways that would’ve been inconceivable a decade prior: gyms, events, rentable flex spaces, amenity apps, and much more.
There is a reckoning to be had among obsolete assets kept afloat for far too long by lazy occupiers and oblivious owners.
The nascent office tailwinds I describe are not going to save shitty suburban office assets, moribund Midwest CBDs, or Class C Garment District buildings without real work and capital invested. And it’s not like the office market in 2019 was particularly healthy with sagging demand for sad assets artificially inflated by WeWork and its peers. A correction was needed.
But it certainly looks as if the rush from office to multifamily over the past five years was a tremendous overreaction — at least on the fundamentals.
The cynic’s view is that winning at institutional allocation has little to do with fundamentals but rather an elaborate game of guessing what asset classes will be en vogue among your peers in five or ten years. Institutional capital flows, after all, drive the bulk of returns on the investment time horizons you care about. It doesn’t matter if the NOI beat your underwriting by 500bps if cap rates are 100bps higher because nobody wants to own the thing. While reallocation (and therefore capital flows) may be influenced by a category’s fundamental performance, it’s a degree or so removed.
While institutional allocators may show increasing concern about the state of multifamily — particularly if they take the long view — they have a bit of a TINA problem. It’s increasingly tough to make money in industrial. Data centers are cool, but it seems a little silly to jump on a bandwagon that’s already 92% of U.S. GDP growth. And you’re not getting back into office.
It’s a reason to be bullish on niche and alternative investment categories. If the “institutional-grade” buckets are all unappealing, might institutions find paths to wander a little off the beaten path, exploring more operationally-intensive categories like healthcare or leisure real estate? Might they be willing to take a little more regulatory risk in, say, build-to-rent or affordable housing?
Real estate fortunes aren’t made by building what institutions want today. They’re made by building what institutions will want in five to ten years. And given how different the world will look by then, there’s a good chance the next institutional asset class is out there waiting.
Brad is the founder and editor of Thesis Driven (www.thesisdriven.com), a publication covering innovation in the built world. Over the past decade he has co-founded General Assembly and Common which generated a combined $500M+ in enterprise value.
