Jim Chanos is betting against data centers

pp76/iStock via Getty Images

Jim Chanos is gambling against legacy data centers as they now go head-to-head with what was once some of their largest clients.

“This is our big short right now,” Chanos said in an interview with the Financial Times.

Chanos added: “The story is that although the cloud is growing, the cloud is their enemy, not their business. Value is accruing to the cloud companies, not the bricks-and-mortar legacy data centres.”

Amazon Web Services, Google Cloud and Microsoft Azure are three of largest and most impactful occupants of data centers around the world and Chanos believes these organizations will opt to create their own centers versus move into an existing one. Therefore, the REITs are to become overvalued.

The FT quoted Chanos saying: “The real problem for data centre Reits is technical obsolescence.”

“Their three biggest customers are becoming their biggest competitors. And when your biggest competitors are three of the most vicious competitors in the world then you have a problem.”

Furthermore, the FT outlined Chanos stating: “People just shrug their shoulders and don’t seem to notice where equity valuations are today versus historically and that there are so many flawed business models. It’s a little bit baffling that no one seems to think they need financial insurance because it’s pretty cheap. It’s another reason to be more cautious — no one is beating down the door of short sellers these days.”

In broader market news, stock index futures bounce back and forth between positive and negative territory on Wednesday as Wall Street looks for direction.

Comments

[Rogier van Vlissingen]

some tangential considerations... www.linkedin.com/...

[WIREDRE]

Secular changes are coming for sure - the undifferentiated REIT players will see pressure and volatility, and technical obsolescence is real. But even chip firms outsource some fabrication, and the idea that ALL firms will go to cloud is not tenable…there’s room and a long runway for DC REITs.

[Trapping Value]

Not sure why any of this is surprising to anyone commenting here. I wrote this on DLR a year back.

seekingalpha.com/...

Their occupancy has declined every single quarter. If they were a mall REIT, with this occupancy they would be going under by now. That will not happen, but that 20X multiple could easily become a single digit multiple.

[Value Investment Club]

His thesis makes sense but those data centers, aren't they responsible for the capex of that equipment in the center? If the big three where to take that in house, they would need a lot more on capex and be complete responsible for the uptime. Seems like the data centers offer a layer of risk transfer. A buffer might be good, keeps the data centers honest to ensure high quality SLA's.

[Skaterdude]

@Value Investment Club It kinda doesn't. Google, for example, has been building their own data centers for years. Ten or so years ago they published an article on trends in disk drive failure rates. They have that much data on operating DCs. FB builds their own servers (yes, I know FB isn't named in the article, but it's illustrative). Is he claiming that AWS doesn't run its own DCs? Odd.

[Up and Away]

@Skaterdude
AWS does and also has done so for years. No one is right all the time!

[Keep it Country]

The 2 quality with Moat data centers REITS are EQIX and DLR. I would be surprised if he would short these 2.

IRM is its own animal with a mix that includes some data center.

3 others that seem more likely to short are DBRG, CYXT and SWCH. Ironically these all have elevated short interest compared to others above.

[cwebjohn]

@Jackson Falls NH No doubt Chanos has already gone after DBRG, which makes little sense considering how much faster they can deploy capital in this environment.

[HammondEhgs]

Microsoft, just to name one, leases entire DataCenters from other providers who own the buildings. Similar to a lease-back agreement, I think. They control who has access, etc. but they don't actually own the structure.

[cwebjohn]

@HammondEhgs Correct. In many circumstances it doesn't make sense for a software company to manage a real estate portfolio...

[Name It]

@cwebjohn Ray Kroc would disagree.

[prema_donna]

@Senator Blutarsky He would've, but fast food companies are actually in the real estate industry. Tech companies are not, so data center REITs have a place. If they were in the real estate industry, they sure wouldn't build their headquarters in the most expensive area in the country.

[Iroha Isshiki]

Interesting but headline is misleading. I thought he was short cloud, but he's short legacy

[Up and Away]

@Iroha Isshiki
He would short his own shorts if he could. Un American.

[PayNoAttention]

@Up and Away His research have served as early red flags on Enron, Wirecard AG, Luckin Coffee, and Hertz. And many others going back to his work at Gilford before founding his firm that exposed Baldwin-United.

Even when he’s been wrong and lost money (He shorted AOL at $8 and covered at $80), he’s been right. How’d that buyout work out for Time Warner?

He’s been wrong (so far) on China and Tesla.

But the idea that putting money behind your measured opinion of a companies future performance is un-American just because that opinion is negative is, to my way of thinking, un-American in and of itself.

[Trapping Value]

@PayNoAttention Yes. Shorting is very hard work and he is a brilliant man who has uncovered a lot of fraud and saved a lot more financial pain.

[Keep it Country]

Respect Chanos and his past calls but this one seems to be an odd ball for him! Did he disclose names?

[DaBan]

I wonder if a sub-Reddit has already formed from this to pump long positions ala Gamestop? I would find that hilarious...

[financeminister]

oh, so this is what happened to my DLR stock today. Well, the reason I bought DLR back in 2013 was because the big short back then during the Irasohn conference was Digital Reality and the short case was made by hedge fund manager Jon Jacobson with a huge slide deck.... I pulled the trigger on buying DLR after his short thesis + taper tantrum bought the shareprice to significant lows and have been holding ever since. Don't think I'll sell this any time soon. Chanos also made a short case against IBM early 2021 and it seemed to make sense. The shareprice was in the ~120ish/shr at the time having a yield of ~5% .... it's one of the few old tech blue chips that have been holding well during this time of volatilty.

[PayNoAttention]

Interesting take. I have enough respect for Chanos and some of his past calls that I always pay attention. He has the resources and knowlege that many of his public calls like this do work out (TSLA being an exception).

I think another way to look at the big 3 building out their own hyperscale farms is that demand is so far outstripping available supply that the existing DC’s can’t meet it.

I’m reminded a little of Amazon building out it’s own last mile delivery service and how many said this was the death knell for UPS and FedEx. It clearly wasn’t. Amazon used its own delivery vehicles AND others. They built it out because their partners couldn’t meet the growing demand.

I think you may see a similar scenario here.

Blackstone buying up QTS gives me some hope that I’m not alone in this belief. They’re pretty smart too.

[cwebjohn]

@PayNoAttention Most definitely is the demand greater than the supply. The TAM is gigantic as computing workloads move from enterprise data centers to the cloud.

[Flannelsg]

Jimmy vs BX, BAM, DLR etc.

[DomL]

he isn't wrong

[RUBYRUBY3]

INTC dominates old iron data farms and makes no chips that are energy efficient to replace them.
The inflection point Dr. Su spoke of.

[cwebjohn]

So does Chanos make a distinction between colo, hyperscale, enterprise and edge dat center?

[thumbsoup]

@cwebjohn Especially edge. Cost of building and operating thousands of small data centers yourself around the globe, or rent space from others. Reminds me of cell phone towers.

[kpmedia]

@cwebjohn I would be surprised if he ever knows there's a difference in DCs, servers, etc. The cloud offerings like AWS are a different animal, and price out far too many people with elasticity utility billing. I don't think he truly understand this area at all, and will be wrong. I think he's getting a bit too giddy from his Coinbase and Bitcoin call, now trying to pick off other tech. Good luck with that (or actually, not).

[cwebjohn]

@thumbsoup Agreed. I also think hyperscale data centers are where these big 3 play in the sandbox together.

[kpmedia]

Chaos is wrong here, doesn't understand DCs.