A version of this article first appeared in the CNBC Property Play newsletter with Diana Orrick. Property Play covers new and evolving opportunities for real estate investors, from individuals to venture capitalists, private equity funds, family offices, institutional investors and large publicly traded companies. Sign up to receive future editions directly to your inbox.
Billionaire Barry Sternlicht, chairman and CEO of Starwood Capital Group, is a legendary real estate investor. Brendan Wallace is an entrepreneur who co-founded Fifth Wall, a venture capital firm that invests in real estate technology and real estate decarbonization. The two first met at the gym. Now, Wallace can say Sternlicht is a mentor and an investor in Fifth Wall, and Sternlicht jokes that Wallace is his trainer.
Together, they gave CNBC Property Play a rare glimpse into how old-school commercial real estate investing is pivoting into a new, technology-driven world order, and how that new world order still relies on lessons learned from the past.
Below are some highlights from the conversation. Edited for clarity and length.
About CRE investment
Sternlicht: We endured a fairly rapid rise in interest rates of 500 basis points, and most of the people who were invested had to pay for that, whether it was real estate yields rising or they weren’t hedged properly. Costs have gone up, expenses have increased, and significant cash flow has been drained from assets that could have been spent renovating them. There is no question that that is long gone and interest rates are coming down. …Jerome (Powell) will step down (as Fed Chairman) next May, and I don’t think anyone will take the job without agreeing to lower interest rates.
I think they should lower their rates. I think the inflation we’re seeing is related to tariffs. It continues. The situation will probably get worse in the fourth quarter as new inventory hits stores and tariffs become impossible to ignore.
WALLACE: The rate hikes that Barry was talking about affected prop tech by definition. Because all the technology companies, all the money-losing companies, went up at the same time. And at the same time, demand from commercial real estate also stopped.
On top of that, a large part of real estate companies’ investments over the past four years have been in decarbonization efforts, so you could say they were anticipating this kind of wave of decarbonization as they sought to comply with new carbon neutrality laws. And I feel like with the election of President (Donald) Trump, it certainly made me feel like I got a four-year hall pass.
About AI and data centers
Sternlicht: There’s probably $20 billion dedicated to (data center) space. I think it’s a different problem than you think. Most of us don’t build until we get a hyperscaler lease. So we get leases from Amazon, Microsoft, Google, Oracle. What we’re looking at right now is the creditworthiness of the tenants, especially Oracle. Because Oracle is doing all these transactions on the back end to (ChatGPT) and Chat is a startup that doesn’t make money and needs hundreds of billions of dollars to grow to the scale they want.
There is no doubt that AI will change the entire world. That change will be much faster than anything we’ve seen before, much faster than the Internet, and certainly faster than the Industrial Revolution. That’s scary to me. In other words, I’m not that satisfied. As I look at… how we spend money and what we can do with AI agents similar to what we’re doing with humans today, and that’s scary for people. I think you have to let people go, right? With a chatbot, you can do the work of 15 people for $36 a month.
Wallace: I was trying to track all of these pretty Byzantine, somewhat incestuous promises that were happening between big tech companies, between digital infrastructure providers. And it’s actually very difficult to track who ends up paying for everything, but ultimately the economy has to pay.
A rigorous way to test whether that makes sense is to look at the amount of AI compute required to fill all the data centers that are in production or announced for production, and assume that tech companies would have to make some profit on top of that to justify it. Right now we don’t, but let’s assume we have to. Take the margin you need and assume that the revenue flows to language models and AI at scale. If you do that calculation, what percentage of the US GDP would that be today? My concern is that it would be 120% of the US GDP.
About your next bet
Sternlicht: We’re actually investing a lot in Europe. Not here. They gave a stimulus package. Prices are low. They don’t actually have inflation. They have no customs duties. Surprisingly, since I’m back from Europe and the Middle East, I can buy anything cheaper in Europe than I can buy here right now.
Wallace: New York City. People overestimate the durability of these changes in the political atmosphere. Within two years of President Trump’s election, we elected (Zoran) Mandani to be New York City manager, and I think these things are working dialectically. In the long run, New York will be very valuable. So if I were a bettor, I wouldn’t need to make a profit for the next four years and I would bet on New York.
