The Office Reset: Five Years After the World Went Remote

The Slowest Crash in Real Estate History

When offices emptied in March 2020 commercial real estate entered the strangest crisis in its history. Everyone could see the demand shock coming. What no one could see was how slowly it would come. Office leases last seven to fifteen years so the damage only affects the leases as each one is renewed and the tenant decides how much space they really need

Five years later most of that arithmetic has finished trickling in. Hybrid work has established itself as the default option for white-collar jobs. Tenants renewing their leases are reducing their footprint often greatly.when the pandemic began. The strictest national measures land a little lower grouping between the mid and upper teens

The exact number depends on who counts and how. What doesn't depend on that is the direction: vacancies on the books delinquencies on office loans also on the books (our CRE status check has the CMBS figure above 12 percent) and the asset class as a whole is worth about a third less than in 2019

The Split That Defines the Market

Here is the most important fact about the office market: averages do not describe almost any real building. Demand did not decline uniformly across all stocks. It fled towards quality

The newer better-located amenity-rich towers what brokers call the trophy level are largely full and charging record rents in major districts. Employers decided that the new office work is attracting workers to return and signaling that the company is doing well and that such work is only worth paying for at the top of the market. A glass tower with a gym a rooftop good lighting and a five-minute walk to public transportation does that job. OneDark 80s slab with cramped elevators is not no matter how cheap the rent is

Everything below the trophy level inherited the whole vacancy problem. The bottom third aging commodity buildings in secondary locations is functionally obsolete: it can't be rented at any rental price that covers the debt weighing on them. People in the sector stopped talking about the "office market" years ago. Now there are two markets one scarce and expensive one plentiful and dying and they share nothing more than the word "office."

How a Building's Value Actually Dies

To understand why the value of an office building can fall off a cliff while that of a trophy tower barely budges it helps to know two elements of real estate math that appear in almost every valuation: net operating income and cap rate

Net operating income or NOI is what a building earns in rent after paying operating costs property taxes insurance utilities administration but before any debt service. It is the number that reflects the performance of the building as a property regardless of how it is financed

the capitalization rate or cap rate is the return a buyer demands to take on that income stream given how risky it appears. Value in the simplest version of the income approach is simply the NOI divided by the cap rate. A stable low-risk building may trade at a lower cap rate since a five percent cap rate means a buyer will pay twenty times the NOI. A risky uncertain building needs a higher cap rate to attract a buyer becausethe buyer wants more performance for the same dollar of income

Here's why the split is more brutal than gentle. A basic building doesn't lose value simply because vacancies increase. It loses value twice: The NOI falls because fewer rentals come in and the cap rate rises at the same time because the building now looks like a riskier bet for any buyer who prices it. Those two effects multiply rather than add and that's exactly why "vacancy went up a little" can turn into "value halved" faster than you think.intuition suggests

Worked Example: NOI, Cap Rate, and the Vanishing Equity

Let me make that concrete with a building that doesn't exist so no one can accuse me of choosing a real address. Call it an illustrative 300,000-square-foot office tower in a secondary center built in the 1980s with the deep floors and small windows typical of that era

In 2019 let's say it's 95 percent occupied. That's 285,000 square feet rented since 0.95 times 300,000 is 285,000. Rent is $50 per square foot per year so revenue is 285,000 times 50 or $14,250,000. Operating expenses taxes insurance utilities andAdministration accounts for about 35 percent of revenue in this example or $4,987,500 since 0.35 times 14,250,000 is 4,987,500. NOI is revenue minus expenses: 14,250,000 minus 4,987,500 equals $9,262,500

Pre-pandemic cap rates for a building like this cluster at around 5 percent. The value is NOI divided by the cap rate: 9,262,500 divided by 0.05 is $185,250,000. Call it 185 million

Now turn to the year 2026 and let the building live out a realistic version of the lower-third story. Occupancy drops to 65 percent worse than the market average because commodity space absorbs all the vacancy burden that the trophy level lost. That's 195,000 square feet leased since 0.65 times 300,000 is 195,000. Rent drops too because a half-empty commercial buildingcompete on price: call it $38 per square foot. Revenue is 195,000 times 38 or $7,410,000

Operating expenses do not decrease with occupancy as income does. Property taxes insurance and a staffing and maintenance base remain largely fixed whether the building is full or half empty. Keep expenses at approximately $4,800,000. NOI is now 7,410,000 minus 4,800,000 or $2,610,000 a decrease of approximately 72 percent withregarding the 2019 NOI

The cap rate moves too and this is the part that intuition tends to overlook. Distressed and half-vacant commodity buildings don't trade at the same 5 percent cap rate as a stable trophy asset. Buyers want more return in exchange for more risk. Let's call the new cap rate 8.5 percent. The value is 2,610,000 divided by 0.085 or roughly$30,700,000

Look what just happened. The NOI fell about 72 percent. The value fell from 185,250,000 to about 30,700,000 a decrease of about 83 percent. The cap rate expansion alone turned a 72 percent income drop into an 83 percent value drop because the same reduced income stream is now capitalized at a less generous multiple. ThatGap the space between the fall in NOI and the fall in value is the cap rate that does its own work in addition to the vacancy

Take the scenario a step further to what looks like a genuinely stressed commodity building. Occupancy drops to 50 percent or 150,000 square feet. The rent achieved drops to $35 per square foot as the landlord chases any tenant willing to sign. Revenue is 150,000 times 35 or $5,250,000. Expenses still largely fixed are around$4,800,000. NOI plummets to $450,000. With a cap rate of 9.5 percent reflecting a genuinely distressed asset the value is 450,000 divided by 0.095 or about $4,700,000

Now suppose still for illustrative purposes that the land under this tower would be worth about $22,000,000 vacated and sold as a development site. In the 2019 and moderate deterioration scenarios the building is worth much more than the land that supports it $185 million and $30.7 million respectively so no one would think of tearing it down. In the severe crisis scenario the building as an income-generating asset is worth around4.7 million dollars less than a quarter of what the bare land is worth. At that point the rational measure for an owner is to not continue operating the building at all. It is to tear it down or sell it for the land because the property is worth more empty than occupied

Case Study: 350 California Street

The clearest real-world example of this mechanism I've seen is 350 California Street an office tower in San Francisco's financial district. Vanbarton Group bought the building in 2019 near the top of the pre-pandemic office market reportedly for around $300 million.a drop that the commercial real estate press at the time pegged at around 80 percent

I want to be careful with the exact figures here because different reports are written differently and I was not present at either closing. What I am confident about is the direction and approximate magnitude: a well-located but aging San Francisco tower lost on the order of four-fifths of its value in about four years which aligns closely with the mechanism of the worked example above. Occupancy fell attainable rents fell with it and the maximum rate a buyer demanded for a half-empty towerin a city in difficulty increased drastically. The three effects reach the same rating at the same time

350 California is also a clear example of why “offices” stopped being a single market. A few blocks away San Francisco's newest and best-appointed towers were still rented to well-funded tech tenants at rents that were barely budging. The building and the trophy tower down the street share a city a zoning code and a name for their asset class. They don't share an investment thesis

Why Conversion So Rarely Pencils

Conversion converting obsolete office spaces into apartments is the photogenic answer to the lower-third problem the one that appears in renderings and press releases about the revival of urban centers. It's real. It's also rationed by physics in ways that have nothing to do with sentiment about urban centers

The artistic term is window-core distance: How far off the ground a point is from the nearest window. Offices built for open floor plans in the 1970s and 1980s often have very deep floor plates meaning the distance can far exceed what a bedroom needs. Residential livability pushes toward rooms that are about 20 to 25 feet from a window. An office tower with a floor 50 or 60 feet deep ends up with a wide center band that cannotthat strip can be converted into hallways storage or mechanical space but it cannot be converted into what is actually rented a bedroom with a window and every square foot that cannot be converted into a bedroom is a square foot that the owner still paid to acquire and transport without generating rent again

Plumbing is the second constraint less visible but just as binding. An office floor typically clusters its bathrooms near the elevator core a handful of risers that serve the entire floor plate. An apartment building needs a kitchen and at least one bathroom in each unit which means placing new vertical risers every 15 to 20 feet along the perimeter of the building punched through a structural frame that was never designed with those penetrations in mind. This is not a renovation.cosmetic.It is closer to a partial reconstruction and that is where the conversion budgets silently exceed their contingency line

Because of both limitations the conversion is only done when the acquisition price is low enough to absorb those costs and that price tends to be close to the value of the land. Here's a simple illustrative version. Let's assume that all the conversion costs new risers exit stairs windows cut into blank walls and replacing systems cost about $600 per square foot a figure in the range that industry feasibility studies commonly cite for a one-story tower.deep. Suppose the finished apartments are worth $650 per square foot once converted. If the developer pays $150 per square foot to acquire the building close to the typical pre-pandemic price for a mediocre asset the total cost is $150 plus $600 or $750 per square foot versus a value of $650. $100 per square foot is lost and the project is never built. If the acquisition price falls to $50dollars per square foot close to the value of the raw land the total cost is 50 plus 600 or $650 per square foot exactly equaling the value of $650. The deal barely breaks even and only a struggling seller gets a buyer there

That's why the conversion wave really began once distressed sales with discounts of 50 to 70 percent off previous values began valuing buildings close to land value. It's not that developers suddenly discovered conversion. It's that arithmetic finally allowed it

The buildings that are most easily converted are the exceptions that prove the rule. One Wall Street in Manhattan the former headquarters of Irving Trust and later Bank of New York Mellon was converted into luxury condominiums that opened in the early 2020s and it worked largely because its floor plates are unusually narrow and irregular for a tower of its size a legacy of its earlier design. Most commodity towers built in later decades are not that shape. They have floor plates.rectangular deep efficient floor tiles that were exactly right for cubicles and exactly wrong for bedrooms

The Other Two Channels: Demolition and Extension

Conversion isn't the only channel that reaches the bottom third and it's not even the most common. Demolition is the non-photogenic channel: some of the worst buildings are worth more as a cleared site than as a going concern which is exactly the arithmetic the worked example above went through. Once the income-producing value of a building falls below the value of the land no renovation will fix it because the building itself has become the liability not the site beneath it. A vacant tower still generates transportation tax insurance minimum security and maintenance costs without generating incomeenough to cover them. Knocking it down stops the bleeding and leaves a marketable and buildable land in its place

The third channel is extension: Lenders extend loans instead of forcing a sale which spreads loss recognition over years instead of forcing it over a quarter. That mechanism and the loan maturity wall behind it is its own story one of the detailed checks on CRE status and maturity wall coverage on this site and I won't repeat that arithmetic here. What matters for this piece is simpler: The extension buys time it doesn't create value. It just changes who has it.the loss and when you finally have to admit it

Between the three channels and a fourth force that no one designed on purpose office construction falling to generational lows the market arithmetic is slowly becoming clear. There is no new supply a demand recovery concentrated almost entirely at the trophy level and the vacant bottom third is gradually reduced through conversion demolition and the eventual end of extend and pretend. It is not a quick process. It was never going to be like this

What It Teaches Beyond Real Estate

Five years watching this reset taught me three lessons that go far beyond real estate

The first is that the duration hides the damage. Long-term leases made the office market look deceptively healthy through 2021 even as the CMBS market which adjusts its prices daily based on actual cash flows was already telling the truth. If you want an early read on a slow-moving asset look for the fastest price change signal available and trust it over the headline number

The second is the compound of second-order effects. Half-empty office buildings became budget crises for cities because office property taxes finance a large portion of urban services in many urban centers. Empty offices during the workday also affect retail and public transportation ridership downtown since fewer commuters mean fewer lunch orders and fewer subway rides. If this becomes a true vicious circle a self-reinforcing spiralsame of lower tax revenues worse services and more flight differs from city to city and remains one of the genuinely open questions I continue to track

The third is that obsolescence is a distribution not a one-time event. The same shock that wiped out demand for basic office space raised the premium over prime space because the underlying demand didn't disappear but rather was reallocated. This is close to the bifurcation pattern that I suspect AI is starting to traverse other adjacent white-collar assets and is worth watching for the same reason: an average that looks calm may be masking a split that isn't

Where the Bifurcation Story Breaks

I've described the division between the trophy office and the commodity office as almost permanent and I should honestly defend the other side because there are real ways this model bends

The first is that cap rates are not purely a verdict on a building. They are also a function of interest rates and the risk appetite of investors in general and both move independently of anything that happens within a specific tower. If long-term rates fall significantly cap rates across the entire office sector could contract again and some of the value destroyed in the above example would return even for a basic building whose leasing fundamentals never improved at all. That would not mean that the problem of thePhysical obsolescence deep floor plates and outdated systems would disappear. It would mean the price would briefly stop reflecting it

The second is that the division is a spectrum not a hard line and capital can grow that spectrum. A well-capitalized owner can buy a tired but structurally sound tower near the base spend real money on lobbies mechanical systems and amenities and reposition it in a silver tier that competes for tenants who leave real trophy space at a discount. Not all basic buildings are doomed to be converted or demolished. Some just need a landlord willing to write the renovation check and adeep enough discount to make it worth writing that check

The third and the one I think is most likely important is politics. Cities are not passive here. Several have offered tax incentives simplified permits or direct subsidies aimed specifically at driving the conversion economy over the line because a renovated building on the tax roll beats an empty one. A subsidy can make my illustrative $750 per square foot conversion compare to a value of $650 even at an acquisition price well above the land value which meansThat the physics I described establishes a baseline not a hard ceiling. The government can advance calculations that private buyers cannot

And there is a version of the bull case for trophy towers that unravels over time. High rents at one level invite new supply to chase that premium developers to aggressively renovate or build again to capture trophy prices and if enough of that supply appears at once the scarcity that maintains trophy rents erodes. The trophy is a moat until it becomes a strategy that everyone tries to copy at once

How I Actually Read an Office Building

If I were handed the annual filing for a REIT with a portfolio of office buildings and asked what to check first I would mostly ignore the occupancy number for the combined portfolio for the same reason I'm wary of the overall vacancy rate for the entire market. A combined number can obscure one wing of the portfolio that props up another

I would go building by building or as close to that as the disclosure allows and ask three questions of each. When was it built and what does that imply about the depth of the floor plate and the distance between the window and the core since that tells me whether this asset could ever be converted if the lease fails. Where does it rank relative to the newer better equipped competition in its own submarket since the offices are actually dozens of hyperlocal markets and the real competitor for a building is the tower three blocks fromdistance not the national average?And what will the expiration schedule of their leases look like over the next three to five years since a building with leases coming soon in a weak submarket is a building whose current lease may not survive the next renewal?

Only after that would you look at the cap rate or implied valuation and interpret a rising cap rate on a specific asset as the market's honest opinion of that building's future cash flows not some abstract risk premium floating in the air. The example worked out earlier in this article is actually that instinct turned into arithmetic: NOI and cap rate together not separately are what tell you whether a building is worth defending or worth walking away from

My honest read - and this is an opinion not a recommendation to buy or sell anything - is that the office sector will continue to look good on paper and get worse in practice until vacancy statistics finally catch up to what a walk through a virtually empty commodity tower already tells you. I find this really difficult to model accurately because much of the value destruction occurs outside the income statement in deferred maintenance and silent lease concessions before it shows up in a cap rate.reported.The way I would really use everything in this article is as a checklist to ask which side of the divide a specific building falls on not as a reason to take a look at the sector as a whole

The Bottom Line

Five years after remote work upended records for office delinquencies vacancies and office loans the overall asset class is worth about a third less than it was in 2019 and the market has permanently split in two: a scarce expensive trophy tier and a moribund commodity tier that is being liquidated by conversion demolition and time. The mechanism is not mysterious once you write it down. Both vacancy andRent affects the NOI cap rates expand on top of that and the two effects multiply into much steeper drops in value than the underlying vacancy number suggests sometimes steep enough to leave a building worth less than the land beneath it. Conversion is only desirable when acquisition prices fall close to the value of the land because the depth of the floor plates and plumbing risers limit the floor depth of an office that can ever be converted into abedroom.The reset is orderly rather than catastrophic precisely because it is slow: building after building loan after loan and their lessons the distrust of averages in divided markets watching for the fastest signal of revaluation and price obsolescence as a spectrum rather than a cliff edge are worth carrying over to whatever is affected by the same pattern next