The best long-term bull case for REITs involves rising replacement costs and falling supply risk, but it’s rarely made in the best way—because it’s kind of a downer, and it involves some awkward twists between real estate and the REIT structure. Replacement cost is also a slightly fuzzy concept in any sector, and it's not a complete REIT bull case by itself. But let me try to lay it out in four parts, which should each be more familiar to generalist investors:

1) We’re talking about a special case of a spiraling cost problem in all kinds of infrastructure and certain kinds of big ticket hard goods, from airports and housing to jet engines and new cars. This is not US-specific, but it’s often worse here for a variety of reasons, and it’s been one of the ongoing puzzles of outsized US growth. We have so much economic dynamism in certain sectors, and along certain competitive vectors… but in other areas, we’ve been sliding into an aftermarket/maintenance economy, with slower economic depreciation in the existing stock.

The HALO-over-SaaS AI trade is one version of this story, right? In fact, “heavy assets, low obsolescence” is a perfect example of how it’s a story that wants an inversion or a positive spin, rather than being presented as a bet on stagnation. As a counterexample, I sometimes describe the residential REITs as the most direct way to bet on NIMBYism, but I can see why they’re not putting that on the cover of their annual reports.

2) Now, in comparison to some other HALO sectors, commercial real estate is easier to analyze. Demand is cyclical in most property types, but it’s broad and granular, and secular demand shifts happen slowly. Much of the value is in the underlying land, which is its own deep market that prices in zoning and entitlement, and discounts potential higher uses. Locational obsolescence in the current use is generally easy to identify, and often so easy that it’s hard to miss.

With the building itself, it’s at least conceptually easy to separate functional obsolescence (this warehouse is inefficient for robotic stacking systems) from positional obsolescence (someone built a fancier office tower down the street). It’s harder to quantify these two factors, and many cheap older assets have elements of both. But all things considered, it’s easier to think about all this with real estate than with many other hard assets, and you won't have to smuggle in as many separate macro/policy assumptions.

3) You can’t always bet on these trends directly by buying and holding public REITs at a deep discount to replacement cost, because they're not patient long-term arbitrage vehicles. They're solving for near- to medium-term earnings growth like most other public companies, which often draws them into similar capex arms races at the high end of the market. This is easiest to see in certain property types, like office towers or A malls or luxury hotels… and demand hits from Covid (e.g. to office attendance or business travel) can make the arms race problem worse.

But it’s not exclusive to those property types, or even to CRE. The generic form is that you’ll buy an asset-heavy stock at a 30% discount to replacement cost, only to watch them sell their most out-of-favor assets at a 50% discount, in order to buy consensus assets at a 10% discount (if you're lucky). There are some REIT-specific plays, but they did not invent this playbook—and even when it works to generate earnings/NAV growth, it's still trading away some of the replacement cost cushion.

4) Fortunately, there’s a counter-twist that turns out to be just as powerful, and a little more CRE-specific. It comes from the Covid-driven step change in replacement costs (as distinct from the creeping long-term trend) and the way it wiped the slate on ground-up development—much of which still doesn’t pencil, even five years later.

What that does is to shift the arms race into existing buildings, where the rising rate environment has lowered valuations and locked up their capital stacks. This is a great competitive environment for REITs, who tend to have lower leverage and more flexible unsecured financing.

For example, imagine you’re an office owner with unhappy LPs that have no interest in funding a renovation or redevelopment, an understaffed loan servicer that won’t call you back, and no incentive to even invest leasing capex in an asset that’s still facing possible foreclosure. It’s easier than ever for the REITs in your market to pick off your best tenants as their leases roll, and the resulting vacancy and expense leakage will make all your problems worse.

It’s that path dependence at the asset level that really kneecaps undercapitalized landlords, right? When you add in loan modifications and other extend-and-pretend behavior, you’ll appreciate why all this "commodity" product is taking so long to recapitalize, and why the opportunity for REITs is lasting longer than you might have expected.

So I sometimes see these public-over-private dynamics boiled down to a “liquidity” advantage for REITs, but that’s really not the best version of the pitch. In some cases it’s missing the point entirely, and making financial markets sound like a free money machine for companies that are trading at deep discounts; liquidity isn’t worth much if it comes at too high a price.

If you want to think about top-down advantages, I’d start with permanent equity: even when the stock is too cheap to issue more, at least they don’t have to worry about redemptions. But it's better to think of a dozen separate advantages that reinforce each other, and vary a little by property type. They're not directly tied to high replacement costs, but they've been amplified by this sharp transition to a lower-supply regime.

OK, you can see why this version of the replacement cost pitch is not being made, much less priced in. It just took me about a thousand words, and I was trying to be as concise as possible. It would also have been hard to convince you of the advantages in point #4 without conceding the arms race problem in point #3, which is another example of how it's an awkward fit for a more standard REIT investment pitch.

Again, discounts to replacement cost are also not a complete investment pitch on their own; they are more of a margin-of-safety story that needs to be supplemented by a growth story, and not every growth story is preserving the margin of safety. We only glanced at this in point #3, but it's a much deeper subject… and that's another reason that replacement cost is often relegated to a single slide or a fuzzy high-level talking point, without the emphasis it deserves.

So hopefully I've restored some of that emphasis, and shown where some of the fuzziness comes from. If you apply these four lenses to the REIT universe, you will also find major differences across property types and individual companies, and there's no question that some are more levered to this "factor" than others. But at the group level, I see it as a growing net positive, and that's why it felt like the right time for this breakdown.

Notes / Links:

For much more on those capital allocation tensions in the third section above, see REITs are not Real Estate (2025) or The Dead Zone (2022). I've also added another old investor note to the samples on the PLRA website, with some further thoughts on the risk of overbuilding the high end.