Chapter 4. Cycles: Everyone Knows, No One Escapes

The Liquidation Order

On January 29, 2024, in the High Court of Hong Kong, Justice Linda Chan read out a brief ruling: the liquidation of China Evergrande Group was ordered. The courtroom held nothing but lawyers in suits, but the sentence liquidated far more than a handful of thirty-story towers. Once the largest landholder in China, with 1,300 projects underway across 280 cities and even its own soccer club, the company had defaulted in 2021, limped along for three more years, and finally shut its doors. In 2021 and 2022 alone it racked up losses of more than $81 billion.¹ Hundreds of thousands of buyers who had already paid for apartments under China’s presale system were left as a line item on a list of unsecured creditors, still waiting for buildings that would never be finished.

At the same moment, on the other side of the planet, a very different drama was unfolding quietly in the US office market. Brookfield, one of the largest real estate managers in the world, was unwinding a large block of office assets near Washington, DC. In the face of structural shifts — remote work, a shrinking federal footprint — one of the industry’s most sophisticated risk managers was cutting loose the very asset class once considered the safest of all: trophy office.² The reasons behind Evergrande’s collapse and Brookfield’s retreat look nothing alike on the surface. But both companies lived through the same experience: the assumption that “this asset was safe yesterday” evaporating overnight.

Evergrande’s liquidation and Brookfield’s sell-off appear, at a glance, to have nothing to do with each other. One is the collapse of China’s state-directed development model; the other is the reshuffling of American offices driven by rate hikes and remote work. The triggers are completely different. But line these two events up against two others from thirty years ago and fifteen years ago, and something strange comes into view. Bangkok, 1997. Las Vegas, 2008. The surface causes were different every time — a currency, a subprime loan, a pandemic, inflation. Dig underneath, and the same thing is buried each time: credit stacked on real estate collateral, arriving at a moment it can no longer be repaid.

Bangkok, 1997: The First Sighting

The story rightly begins in Bangkok. In the early 1990s, Thailand was known as one of the “Asian Tigers” — an economy growing 8 to 9 percent a year, flooded with foreign capital, and real estate was the easiest channel for absorbing it. Cranes filled the Bangkok skyline. Banks competed to lend to developers, who used the land they bought with that money as collateral to borrow more and buy still more land. No one wrote the assumption that “this growth might someday stop” into a single pro forma. The Thai baht was pegged to the dollar, and trusting that peg, Thai companies borrowed in dollars and poured the money into Thai real estate.

In July 1997, the Bank of Thailand could no longer afford to defend the baht and let it float. The currency collapsed almost instantly. The trouble was that Thai corporate debt was denominated in dollars. Once the exchange rate was cut roughly in half, rental income earned in baht could no longer cover even half of the dollar-denominated debt. Half-built concrete skeletons stood like ghosts across downtown Bangkok for years. The crisis spread immediately to Indonesia, South Korea, and Malaysia. The visible trigger was the exchange rate. But the reason the currency collapse gouged the real economy so deeply, and for so long, was the excess credit stacked underneath it in the form of real estate loans. Had it not been the exchange rate, something else would have pulled the trigger. The gunpowder was already piled high.

Las Vegas, 2008: Same Gunpowder, Different Trigger

Eleven years later, the gunpowder piled up again, this time on the other side of the Pacific. The form of the collateral had changed. It was no longer commercial development loans but subprime mortgages, sliced and repackaged into securities. In the early 2000s, the belief that US home prices “never fall” was practically a religion. Loans covering 100 percent of a home’s value were sold to borrowers with poor credit and no verification of income, and those loans were sliced up again on Wall Street and sold to pension funds and banks around the world.

Las Vegas was this frenzy in miniature. From the early 2000s, new subdivisions covered the desert south of the Strip like a rising tide. Real estate brokerage became one of the city’s largest industries, and “flipping” — buying a house and reselling it within months — became a side hustle for ordinary office workers. From its 2006 peak to its 2012 trough, the median home price in Las Vegas fell by nearly 60 percent. A large share of the city’s housing stock ended up underwater, worth less than the balance still owed on it.

The trigger was not an exchange rate but the collapse of subprime securities. The underlying structure was identical to Bangkok’s. On top of the assumption that real estate values would rise forever, credit was piled up without any real verification of the ability to repay, and the moment that assumption broke, leverage began working in reverse. In Bangkok, the exchange rate collapsed and dollar debt exploded; in Las Vegas, home prices collapsed and loan balances overtook asset values. The mechanism spoke the same language in both places: a decline in collateral value, amplified through leverage.

The fact that leverage is an amplifier becomes clearer with numbers. Say an investor who buys a building with cash alone earns a 9 percent return on assets (ROA). Add a loan covering half the purchase, at a 7 percent interest rate with an 8 percent mortgage constant (the debt-service burden), and the return on equity (ROE) doesn’t stay at 9 percent — it jumps to well into the double digits, because the denominator shrinks from total assets down to the equity actually invested. The problem is that this lever doesn’t care which way it points. The moment rents soften, vacancy rises, or collateral value itself collapses — pushing the asset’s return below the cost of servicing the debt — that same lever amplifies losses by exactly the same multiple. When home prices in Las Vegas fell 20 percent, the equity of an owner who had borrowed 80 percent of the purchase price evaporated, in theory, entirely. When the baht was cut in half in Bangkok, a developer’s equity behind dollar-denominated debt disappeared by the same arithmetic. At the top of the cycle, leverage is a friend. The moment the cycle turns, the same leverage becomes the rope that tightens first around the borrower’s neck.

2020, and 2021–2026: The “This Time Is Different” Illusion, Again

The trigger in 2020 was a virus. Pandemic lockdowns emptied offices around the world, and central banks slashed interest rates to historic lows to prop up their economies. As rates fell into the 3 percent range in 2020 and 2021, the cheap money flooding the system poured into real estate once again.³ This time the money was absorbed not by offices and retail but by logistics centers, residential property, and a newly ascendant asset class: data centers. The narrative that “zero rates are the new normal” spread among investors. Real estate was redefined as a safe asset that paid a better yield than a savings account.

In 2022, central banks raised rates at the fastest pace in history to fight inflation. Capital structures assembled during the zero-rate years — buildings bought with cheap debt and priced at low cap rates — went out of fashion overnight. Once government bonds began paying 4 to 5 percent with essentially no risk, the one thing real estate had going for it — a “safe yield” — stopped being special. As of 2026, industry forecasts put the repricing of commercial real estate values at 20 to 25 percent below peak.⁴

Up to this point, the pattern is familiar. Bangkok had a currency; Las Vegas had subprime; this time it was interest rates. The trigger changes every time, but what stands behind it never does: excess credit built up during a period of low rates and optimism, and the moment that assumption is betrayed, leverage runs in reverse. One academic review reconfirms that most of the major financial crises of recent decades trace their origins back to real estate bubbles.⁵

This cycle has two features that set it apart from the previous three.

The First Anomaly: Demand That Doesn’t Come Back

The crises of 1997 and 2008 were crises that “recovered.” Bangkok’s ghost towers eventually filled up. Las Vegas’s underwater homes eventually regained their value, roughly a decade later. When credit dried up, the market froze; when credit flowed again, the market thawed. It was a cycle in the true sense.

The 2021–2026 office cycle is different. What the pandemic left behind was not a temporary shock but a permanent change in human behavior. As of 2023, US office occupancy remained at roughly half its pre-pandemic level, and the research firm Capital Economics forecasts that US office values will not recover to their pre-pandemic level until 2040.⁶ This isn’t vacancy that refills once the economy improves. It means people no longer need the act of commuting to an office nearly as much as they once did. The buildings still stand, perfectly intact. But a large part of the reason those buildings needed to exist has quietly disappeared.

As a result, this cycle’s repricing is not a wave that falls and then rises again — it is a structural bottleneck, in which the loan structures assembled during the low-rate years must roll over en masse. Across 2025 and 2026, commercial mortgages coming due are estimated, depending on the source, at roughly $1.8 trillion.⁷ It’s the bill that arrives after the party is over. And this cycle is widening the gap — to an unprecedented degree — between the institutions that can afford that bill and the individuals and smaller players who cannot.

The Second Anomaly: There Is No Longer Just One Clock

The epicenter of the 1997 Asian financial crisis was Asia. The epicenter of the 2008 subprime crisis was the United States. And Asia, ground zero in 1997, actually came through 2008 in comparatively good shape — the painful reforms of 1997 had toughened its banking systems in the meantime. The last two cycles moved their epicenters around, but they still converged on a single global clock.

Not this time. While the West went through its rate cycle, China went through its own deleveraging of real estate on an entirely independent timetable. Evergrande’s collapse unfolded on a different schedule than the American rate hikes, and for different reasons. If the West’s trigger was inflation and tightening, China’s trigger was the “three red lines” policy — leverage-ratio regulations the government introduced in 2020.⁸ As a result, as of 2026, global real estate is no longer running on one cycle but several clocks, each ticking to its own rhythm: Europe bottoming out even as the US office market is still hunting for a floor, and China moving through an entirely different phase altogether.

The 18-Year Clock, or the Myth

Here it’s worth pausing on a different kind of clock. In 1933, the economist Homer Hoyt, studying land values in Chicago, found an odd regularity: real estate prices tended to rise for roughly fourteen years and then collapse over four, a pattern that had repeated for nearly a century. The British economist Fred Harrison later refined this observation and gave it a name: the “18-year real estate cycle.” Line up the US troughs, and the clock aligns with unsettling precision — 1933, 1952, 1970, 1990, 2008. Harrison used this pattern to publicly predict the 2008 crash years in advance, which drew considerable attention at the time.

Whether this theory describes a genuine underlying law or is simply a plausible-sounding string of coincidences remains a matter of debate even among economists. But if the clock holds, the next trough should fall somewhere near 2026. And indeed, as of 2026, parts of the real estate industry are once again discussing whether “the 18-year cycle is coming to a close.”⁹ The case for scientific precision is thin. What makes this number meaningful isn’t its predictive power — it’s that it tells the same story every single time. Roughly once a generation, humans borrow, build, and convince themselves that real estate will rise forever, and then forget all over again.

Crossing a Border Makes the Cycle Crueler

The same credit cycle lands very differently depending on whether the person caught inside it is a citizen or a foreigner. For Evergrande’s or Las Vegas’s domestic buyers, the downturn meant a loss of wealth — but at least they had their own courts, their own language, their own politicians to turn to. Foreign investors who climbed onto the same cycle across a border have no such cushion.

While the Turkish lira lost more than 80 percent of its value against the dollar after 2018, European investors who had bought condos in Bodrum and Antalya had to watch their properties stay intact while their principal simply vanished. Rental yields looked fine in local currency, but the moment those earnings were converted back into euros, the lira’s decline swallowed the rental gains several times over.¹⁰ Around the same period in Egypt, the Egyptian pound lost roughly two-thirds of its value (about 68 percent) against the dollar since March 2022. Marketing materials touted nominal price increases of more than 100 percent for units in the New Administrative Capital, but the real value, converted into dollars, was actually being eroded.¹¹ In both cases, the outcome was driven far less by the cycle in the real estate market itself than by the cycle in the country’s currency. The moment an investor crosses a border, they are riding two cycles at once — property and exchange rate. When both turn in the same direction, the losses don’t simply add together. They multiply.

What People Say at the Top

The most striking commonality across thirty years of repetition isn’t found in the numbers — it’s in the lines people say. At every peak, most market participants said nearly the same sentence, in different languages.

In Bangkok in 1996, real estate seminars rang with the line “Thailand will become Asia’s fourth tiger, and this growth has no end.” In Las Vegas in 2006, a statistical illusion — “US home prices have never fallen nationwide at the same time” — was formally written into underwriting models as an assumption. At real estate conferences in New York and London in 2021, the line that drew applause was “ultra-low rates have become structurally permanent; inflation is now a relic of the past.”¹² All three statements sounded reasonable at the time they were made. All three were turned completely upside down within a few years.

The fact that this pattern has repeated more consistently than any other indicator across thirty years suggests that the real way to forecast a cycle might not be a statistical model at all, but psychology. The moment a market begins declaring, openly, repeatedly, and with total confidence, that “this time it’s structurally different” — that may be the single indicator that sits closest to the top of the cycle.

Back to the Revolving Door

Even so, this cycle cannot simply be told as a story ending at the edge of a cliff. Evergrande was liquidated, but the global real estate system did not collapse. When the $1.8 trillion wall of maturities actually arrived in 2025 and 2026, what happened was not a repeat of 2008. A new ecosystem — private credit — stepped into the space banks had vacated, and a large share of building owners whose loans came due were able to refinance on new terms. It was not a crisis. It was a revolving door.

It’s worth looking more closely at how that revolving door came to function. The lessons of 2008 stayed etched into the banking system, and regulators around the world subsequently tightened bank capital requirements substantially. As a result, by the time the 2021–2026 rate cycle actually hit, a large share of commercial real estate lending had already migrated off bank balance sheets and into the more loosely regulated world of private credit. The system avoided a 2008-style chain reaction that shakes the entire banking sector — but the risk was instead dispersed into a private market that is far less transparent. Building owners facing maturities were able to quietly roll into new terms not because regulation disappeared, but because the type of capital absorbing the risk had itself changed. No one can yet say with confidence whether this new ecosystem will play the same cushioning role in the next cycle — because this is the first time the private credit market itself has faced a real stress test since 2008.

The way this cycle’s recovery arrived also sets it apart. Past recoveries were driven mainly by macro policy — rate cuts, bailouts. This time, even as offices were collapsing, capital rapidly reclassified itself into emerging sectors — logistics, data centers, senior living — splitting commercial real estate statistics into extreme winners and extreme losers. The structural signature of this phase wasn’t a broad rebound across all of real estate at the bottom of the cycle; it was a K-shaped exit, with some sectors permanently exiting the stage and others posting record highs. That story continues in the next chapter, on the death and rebirth of the office.

Rule of the Game

The real estate cycle is not an interest-rate cycle. It is a credit cycle. The trigger changes every time — a currency, a security, a virus, inflation — but underneath it, there is always the same thing: excess credit stacked on real estate collateral.

The most accurate indicator of a cycle isn’t a statistic. It’s a line of dialogue. The moment the phrase “this time it’s structurally different” starts drawing applause in a conference hall, that moment sits closer to the top than any indicator ever could.

When credit dries up, markets freeze — but they don’t always collapse. Sometimes they turn into a revolving door instead. What decides the difference isn’t the size of the crisis. It’s how quickly a new lending ecosystem fills the space the banks vacate.

---

Sources

Footnotes

1. Statista, “The Staggering Losses of the Chinese Property Crisis Emerge” — cumulative net losses of roughly $81 billion across 2021–2022 for Evergrande and other Chinese developers; CNN Business, “Evergrande, symbol of China’s property crisis, heads to liquidation” (2024) — the Hong Kong High Court’s liquidation order of January 29, 2024. ↩

2. Compiled from industry reporting on Brookfield’s sale of office assets near Washington, DC. See Chapter 2 for a detailed analysis of the institutional capital exiting this sale and the family offices and opportunistic funds stepping in to buy. ↩

3. Deloitte Insights, “2026 commercial real estate outlook”; Morgan Stanley, “Real Estate at an Inflection Point” — policy rates falling to historic lows (roughly 3 percent) in 2020–2021, followed by a sharp subsequent tightening. ↩

4. PwC/ULI, “Emerging Trends in Real Estate: Global 2026”; Principal Asset Management, “2026 Inside Real Estate Outlook” — estimated 20–25 percent repricing of commercial real estate values from peak. ↩

5. MDPI Real Estate, “Cycles, Trends, Disruptions: Real Estate Centrality on the GFC, COVID-19, and New Techno-Economic Paradigm” (2023) — an academic review reconfirming that many of the major financial crises of recent decades originated in real estate bubbles. ↩

6. Capital Economics forecast, cited via Deloitte Insights and others — US office occupancy at roughly half its pre-pandemic level as of 2023, with office values not projected to recover to pre-pandemic levels until around 2040. ↩

7. Mortgage Bankers Association (MBA) tabulation — combined commercial mortgage maturities across 2025–2026 of roughly $1.8 trillion (approximately 7,000 assets); some private research estimates put the figure above $2 trillion. ↩

8. Background on the policy trigger behind China’s real estate crisis (the “three red lines” leverage regulations introduced in 2020) — Council on Foreign Relations, “Does Evergrande’s Collapse Threaten China’s Economy?”; Wikipedia, “Chinese property sector crisis (2020–present)” (compiled from public reporting for cross-verification). ↩

9. The “18-year real estate cycle” theory, refined by Fred Harrison and others, aligning US troughs in 1933, 1952, 1970, 1990, and 2008; Norada Real Estate, “What is the 18-year Real Estate Cycle?”; BiggerPockets, “The ‘18-Year Real Estate Cycle’ Ends in 2026 (What Now?)” — cited for the industry discussion as of 2026. Note: the methodological precision of this framework is low, and it is presented here as a popular observational frame rather than an academically validated law. ↩

10. Compiled from reporting by Property Guides, The Luxury Playbook, and others — the Turkish lira’s decline of more than 80 percent against the dollar since 2018, and roughly $8.7 billion in foreign capital outflows in 2025 alone. ↩

11. Global Finance, “Egypt Devalues Currency, Raises Interest Rates”; exchange-rate data (XE, TradingEconomics) — from roughly 15.65 EGP to the dollar in March 2022 to roughly 49.6 EGP to the dollar in mid-2026, a decline of about 68 percent against the dollar. (Note: 8.88 EGP/USD reflects the fixed exchange rate in place before the November 2016 shift to a floating regime, a separate period from the post-2022 devaluation discussed here. ) ↩

12. These three statements are reconstructions of narratives that circulated widely at each cycle’s peak; they are not verbatim quotations attributed to any specific individual. ↩