Office real estate stocks tumble as AI disruption casualties in the stock market grow by the day

Published Thu, Feb 12 202612:52 PM ESTUpdated Fri, Feb 13 202610:43 AM EST

Key Points

Commercial real estate brokers are selling off for a second straight day, with CBRE tumbling 12.8%.
CBRE's drop is especially alarming given that the only other times the stock has tumbled further was during Covid and the height of the global financial crisis, according to Oppenheimer.
That sell-off reflects a grim mood as of late in the market, which has rotated sharply out of those companies most exposed to AI disruption — first in software, then in financial firms.

Real estate stocks have become the latest victim of the artificial-intelligence threat. Commercial real estate brokers are selling off for a second straight day. CBRE closed nearly 9% lower on Thursday, after tumbling more than 12% earlier in the session. Jones Lang LaSalle closed down 7.6% and Hudson Pacific Properties shed nearly 4%. In addition, Newmark and BXP both slipped more than 4%, while SL Green Realty dropped about 5%. "We believe investors are rotating out of high-fee, labor-intensive business models viewed as potentially vulnerable to AI-driven disruption," Jade Rahmani, an analyst at Keefe, Bruyette & Woods, said in a note Wednesday. CBRE YTD mountain CBRE year to date That sell-off reflects a grim mood as of late in the market, which has rotated sharply out of those companies most exposed to AI disruption — first in software, then in financial firms — for more defensive sectors such as staples. On Thursday, trucking and logistics stocks also tumbled on the release of an AI freight scaling tool. Shares of C.H. Robinson Worldwide and RXO plummeted 20% and 25%, respectively. Shares of J.B. Hunt Transport Services slid more than 6%. Now, investors are on the lookout for what sector will be the next domino to fall, and how long any panic selling can last. AI disrupting employment Commercial real estate has been under pressure for some time, as higher interest rates and the rise of remote and hybrid work in the wake of the pandemic cratered demand for office space. Investors worry that AI could sound a death knell for the sector. That point was driven home in an essay that went viral earlier this week in which OtherSide AI co-founder and CEO Matt Shumer said entry-level, white-collar jobs will be gutted thanks to AI. The impact will be bigger than Covid, he wrote. The essay garnered 30 million views in 24 hours, Shumer claimed. Those remarks follow Elon Musk's comments on a podcast last week in which he said that office towers once filled with workers will one day be replaced with AI. "Corporations that are purely AI and robotics will vastly outperform any corporations that have people in the loop. Computer used to be a job that humans had. You would go and get a job as a computer where you would do calculations. They'd have entire skyscrapers full of humans, 20-30 floors of humans, just doing calculations. Now, that entire skyscraper of humans doing calculations can be replaced by a laptop with a spreadsheet," Musk told the hosts of the "Dwarkesh Podcast" last week. "That spreadsheet can do vastly more calculations than an entire building full of human computers," Musk added. SLG YTD mountain SL Green, YTD The rout in real estate stocks comes on the heels of several other sectors being dragged down by AI disruption concerns. Software stocks took a hit earlier this year after Anthropic's latest AI mode l appeared able to allow businesses to do legal work and build programs for which they would otherwise pay an expensive license. Then wealth management stocks dove after the launch of tech platform Altruist's new AI-powered tax planning tool that promises to do the work "within minutes." Fears overblown? Even so, many investors expect that the recent concerns could be overblown. Indeed, in spite of all the noise, fundamentals in real estate remain strong, Rahmani noted. "While the threat of technology disintermediation is not new to the industry, the current sell-off may overstate the immediate risk to complex deal-making, even as the long-term AI impact remains a 'wait-and-see,'" he wrote. In fact, CBRE reported an earnings beat on Wednesday for its fourth quarter and issued strong guidance for the full year. Its core earnings came in at $2.73 per share, topping the consensus estimate of $2.68 a share, per FactSet. The company expects core EPS to come between $7.30 to $7.60, versus the $7.39 expected from analysts. CBRE CEO Bob Sulentic pushed back against the perception that the company's core businesses will be disrupted by AI, saying that the firm has built cost-effective AI tools to aid, but not disrupt, the work of its brokers. He added that much of the transactions CBRE oversees are complex, requiring the firm's deep knowledge and breadth of relationships in the field. "We've become quite confident that that business really is driven by this strategic creative thinking that our brokers do," Sulentic said during the company's earnings call. "And we think that's going to continue to be the case, and we haven't seen any evidence to the contrary." Barclays analyst Brendan Lynch is sticking with his overweight ratings on CBRE and Newmark and would buy the weakness. "We see the harsh sell-off among the group as inconsistent with their earnings profiles," he said in a note Wednesday. "We do not dismiss this risk, but note that thus far AI has been a net job creator," he wrote. "Further, CRE servicers stand to benefit, like many other companies, from both revenue growth opportunities and cost synergies." However, there could be long-term implications for businesses that don't shift from using AI as another tool in their toolboxes to a core operating infrastructure of new business models, said Macquarie strategist Thierry Wizman in a note Thursday. For instance, for financial services and real estate companies, outcome-driven AI agents would conduct all the end-to-end workflow, replacing the human-led ones, he said. "[F]or companies that are slow to adopt, or have built customer models based on costly human-level discretion and interaction, that transition may be fatal," he said.