CRE Analyst’s Post

Here are the trends and events that we think will define commercial real estate in 2024.

Most renown lease termination of all time?

Bottom-third office buildings, vacancies at hand. Defaults threaten banks, like Brody on Homeland. ---- Hiding in plain sight ---- We know where the struggling office buildings are. 90%+ of all vacant office space is concentrated in a third of the office stock. So the worst buildings have 3x their proportionate share of vacancy. These buildings are likely toast. Most of them may only be worth the land they sit on. ...they're a ticking time bomb in a way. Many have long fuses, but ticking nonetheless. ---- $100B of losses? ---- It's not hard to come up with $100 billion of losses from these ticking time bombs. Back-of-the-envelope assumptions... -- Total collateral value = $400 billion (2.8 bil sf x $140/sf) -- 70% average LTV -- $280 billion of bottom-third office loans at banks -- 50% default rate = $140 billion -- 75% LGD = $100 billion of losses -- 6% of total outstanding bank loans Bank reserve ratios are generally between 1-2%. However, maybe there's a mitigating factor. ---- Slow burn ---- It's extremely feasible that office distress will play out over several years. Loan maturities are relatively staggered. The cap rate shock has (to date) been much more severe than the property income shock, leaving relatively favorable debt service coverage. ---- Sidelined ---- Even if this plays out over a long time, lenders and owners related to these bottom-third buildings could easily be sitting on $200+ billion of unrealized losses. With or without insolvency, these time bombs will likely sideline many lenders and buyers as the market thaws. ---- What to watch ---- Who owns or lends on these bottom-third buildings? Who doesn't own or lend on these bottom-third buildings? PS - Would you rather buy (i) a bottom-third building for pennies on the dollar, (ii) a middle-third building for a 50% discount, or (iii) a top-third building for a 25% discount?

If you use one of these… What career advice do you have for those who don’t? They’ve never been through a real estate recession.

"The first step in solving a problem is to recognize that it does exist." Put yourself into the shoes of a perpetual real estate vehicle over the last few years. These vehicles generally fall into one of two camps... ---- Dracos ---- Asset manager: "Our property values are in free fall." Investor relations: "Let's issue a press release saying our appraised values are holding up. Concerns are overblown." Fundraiser: "We have a problem. No one will call me back until they think our values have stopped falling." Acquisitions: "We're starting to see good acquisition opportunities." Chief investment officer: "Acquisitions? We can't buy anything until we regain investor confidence. We need to sell a few of our most liquid properties to generate cash." ---- Potters ---- Asset manager: "Our property values are in free fall." Investor relations: "Let's issue a press release explaining that, despite capital markets pressures, our balance sheet is sound." Fundraiser: "Investors appreciate our candor but don't want to get back into the mix until the transaction market starts to thaw." Acquisitions: "We're starting to see good acquisition opportunities." Chief investment officer: "Since we've taken our lumps, we have dry powder. Put in a fair offer." ---- Commentary ---- Publicly traded REITs are the Potters. As this chart shows, they've taken their lumps and have access to capital. The strongest REITs are in especially good shape. They're natural buyers. Private perpetual vehicles (e.g., non-traded REITs, open-end core funds, open-end core+ funds, etc.) are the Dracos. They can't help it. The slow but certain re-pricing has shaken confidence, exacerbated aggravation, and sidelined them. They're natural sellers. These two profiles are much more connected than many people think.

A good time to own data centers?

A generational CRE hangover... ---- Three fundamental shifts ---- 1. Less excess return The 10-year Treasury rate was 6.7% in the 1990s. It fell to 4.5% in the early 2000s, to 2.4% in the 2010s, and sub-2 % during the pandemic and subsequent recovery. Real estate return requirements also fell: 10%+ in the 1990s, 8% in the early 2000s, 7% in the 2010s, and 6% during the pandemic recovery. ...but the spread between treasury yields and real estate returns was the highest on record as Treasurys reached their trough. CRE returns were 0.6x Treasuries in the 1990s and peaked between 2-3x Treasuries when the 10 Year approached 0%. 2. Higher return requirements Back to the future? The dust hasn't settled, but it's clear that the 'free money' days of real estate are over. The 'new' reality is reminiscent of the 1990s. Buyers are demanding higher returns, pushing down values, which hurts sellers. 3. More dispersion The difference between winners and losers, within CRE, is growing. In the 2000s, there was very little difference between the highest and lowest yielding real estate sectors. But that difference has grown from about 10% to about 50%. ---- Good news: No existential crisis ---- Despite a steady flow of real estate 'doom loop' headlines, the industry does not face an existential crisis. Why is this so difficult for pundits to believe? Perhaps because CRE has been whipsawed by crises since the 1980s: Extreme overbuilding, widespread excess debt (S&L), and excess debt again (GFC). But declining values don't necessarily equate to a crisis. Sometimes, it's just a cycle. Normal cycle declines stay in the fairway (i.e., 25% ish value declines), while existential declines threaten the fundamental structure of the industry. The industry, as a whole, is in the fairway. PS - Another bit of good news: This framework is yet another indicator suggesting that real estate players will increasingly get paid for skills (vs. simply being in the market). This will likely be painful for some ('why am I not getting a call back') but opportunistic for others ('we need someone who can do [xyz].')

A big step toward the industry's new normal? Yesterday, KKR announced it has agreed to buy $2 billion of apartments from Lennar. Two months ago, Blackstone agreed to pay $10 billion for AIRC. We think these trades suggest a new watermark for higher-quality apartments between 5% and 5.5% (cap rate). Two implications... 1. Running out of of altitude: Many apartment buyers purchased properties during the Covid rebound at cap rates well below 5-5.5%, and their yields are likely worse now due to sluggish rents and higher opex. A large share of these buyers purchased lower-quality assets with fix-and-flip strategies and short-term, floating rate debt. This plane ran out of fuel a few years ago and has been gliding toward a crash since rates began to spike. ...but these borrowers are quickly running out of altitude. 2. Thriving survivors: Those who weren't loading up on class B and C apartments with excess floating rate debt are back in the game and aren't as distracted by interest near-term maturities, loan paydowns, and rate cap renewals. These dichotomies are unquestionably simplistic and ignore nuances. ...namely, that many firms will have properties in both categories. However, we think they frame a positive turning point for the industry. Apartments remain the most financeable assets in commercial real estate, and these big trades provide incremental pricing clarity. The fix-and-flip hangover is just beginning, as are the opportunities for well-capitalized buyers. Either way, clarity and increased activity represent a meaningful step toward the industry's new normal. Chart detail: One large apartment REIT's FFO yield vs. market cap rates since the Covid recovery. The significant gaps shed light on pending winners and losers.

Would you rather buy office buildings at a 10% cap or NVIDIA at a 1.5% cap? A week ago, we pointed out that NVIDIA has added 1.5x the entire REIT universe's market cap so far this year. i.e., that one tech stock has added more than the equity in all 193 REITs during the first six months of 2024. [As of last week's peak valuation, NVIDIA's market cap equated to about $100M per employee.] But over the last three days, NVIDIA has shed about $430 billion in market cap. ...which (by our count) is greater than the losses in retail, industrial, and hotel since the recent interest rate trough/value peak. ...is approaching the total loss of value in the office sector. ...and, surprising to many, multifamily remains significantly ahead in terms of value loss. Why? Super low yields were disproportionately affected by rate spikes, and it's the largest real estate subsector. But which topic has captured more negative headlines: real estate doom loops or an NVIDIA/AI bubble?