Every major real estate crash followed the same pattern

For over two centuries, real estate has followed a rhythm so consistent that multiple researchers - working independently, decades apart - arrived at the same number: approximately 18 years.

Homer Hoyt documented it in 1933 studying a century of Chicago land values. Roy Wenzlick charted it from 1795 to 1974 and calculated an average of 18.33 years. Fred Harrison traced it 300 years in UK land prices. Phil Anderson refined it to exactly 18.6 years.

The pattern is remarkably stable.

Harrison and Anderson agree on the internal structure: roughly 14 years of rising prices followed by 4 years of decline. Anderson subdivides the upswing further, identifying a crucial "mid-cycle wobble" roughly halfway through:

Anderson was emphatic that COVID was not a random shock but the mid-cycle slowdown arriving on schedule - the same structural role played by 9/11 in 2001, the Volcker rate shock in 1981, and the Cuban Missile Crisis in 1961. After each mid-cycle wobble, the most speculative phase of the boom begins.

Fred Harrison predicted the 1989 crash in his 1983 book, and the 2008 crash in his 2005 book. Phil Anderson called the GFC and predicted recovery would begin in 2010 when virtually no one else did. Both point to the same year.

Harrison's "Winner's Curse" concept describes the final 18–24 months of the cycle: credit becomes dangerously loose, bidding wars escalate, and the "winner" of each auction is actually cursed - they've overpaid at the worst possible moment. Anderson calls this the "melt-up" phase and points to frenzied speculation in crypto and meme assets as a hallmark.

"This is the worst time to gear up. Sit on cash. The opportunity to buy again will come - maybe around 2030."

- Phil Anderson, May 2025

Warren Buffett appears to agree. Berkshire Hathaway has been building cash for 12 consecutive quarters - reaching $381 billion - while selling its real estate brokerage HomeServices of America. He is behaving exactly as a disciplined investor would at the top of an 18-year cycle.

The 18-year cycle is not a mystical pattern. It is the mechanical output of how land markets interact with credit systems, construction lags, and human memory. The same structural forces have reproduced it for over two centuries.

1. Land is fixed, but demand is not. Unlike manufactured goods, land cannot be produced in response to rising prices. Increased demand simply raises the price - which raises expectations of further price increases, drawing in more speculative buyers. The self-reinforcing loop that inflates every bubble begins here.

2. Credit amplifies the spiral. Land comprises the majority of collateral for all bank lending. Rising land values allow borrowers to pledge more collateral, enabling larger loans, which fund higher bids, which push land values further up. The run-up to 2008 was the clearest modern demonstration: mortgage debt and land values rose in lockstep until neither could support the other.

3. Construction lags by 5–7 years. Developers respond to rising prices by building - but the planning, permitting, and construction process takes years. Supply arrives after the market has already peaked, colliding with weakening demand and creating the glut that drives the bust.

4. Human memory is exactly one cycle long. The generation that lived through the last crash (roughly 18 years prior) is still active but has been gradually displaced in markets by participants who have no experience of a downturn. Institutional memory of risk fades. Lending standards loosen. The cycle resets.

These 4 forces operate independently of policy, technology, or any external shock. Wars and recessions can delay or compress the cycle - World War II suppressed the expected boom of the 1940s; the 1970s inflation distorted the timing slightly - but the underlying pressure always reasserts. Harrison's core thesis, supported by 300 years of UK land price data, is that the cycle is not a law of nature but a product of institutional design: specifically, the taxation of labor and capital rather than land, which leaves land rent as the primary vehicle for private wealth accumulation.

The affordability data is consistent across markets. At each cycle peak, price-to-income ratios reach levels that are arithmetically unsustainable. In high-risk cities at the current peak, prices have risen by nearly a quarter over the past five years, while rents increased by about 10% and household incomes by only around 5%. The gap between prices and fundamentals is not a new observation - it is the defining signature of the late-cycle Winner's Curse phase.

What makes the 18-year cycle unusually credible is not one researcher's claim - it is that multiple economists, working independently across different countries and centuries, arrived at the same number.

The cycle has been baked into the US economy right from the start. US land sale volume data going back to 1800 shows a spike in land sales every 18 years - with a bust following every single peak. The consistency across two centuries and multiple independent researchers is the primary reason cycle analysts treat the pattern as structural rather than coincidental.

The honest case against the 18-year cycle as a crash predictor: the post-2008 cycle has been distorted by unprecedented central bank intervention. Zero-interest-rate policy from 2009 to 2022 extended the recovery phase and suppressed the normal credit clearing process. Anderson cautions that while rate cuts may temporarily fuel speculation, they cannot prevent the crash that inevitably follows the cycle's peak. "The Fed follows the cycle. It does not lead it."

A second objection: the 2026 peak call relies heavily on the 2008–2012 trough as the cycle start. Anderson and Harrison both agree the last trough was around 2009, meaning at 16+ years in, we are approaching the terminal phase. But if the true trough was 2012 (when many markets actually bottomed), the peak shifts to 2028–2030. The directional call remains the same; the precise timing is debatable.

Multiple independent frameworks - cycle theory, market structure, institutional positioning, and affordability data - converge on the same window. This is not one signal. It is a cluster.

The cycle's terminal phase - what Anderson calls the Winner's Curse - is characterized by speculation that extends well beyond real estate into adjacent asset classes. This is not a side effect of the cycle; it is a diagnostic marker of where we are in it. Each prior terminal phase had an equivalent: 1920s stock mania, 1980s junk bonds, 2005–2007 structured credit. Today it is crypto and meme assets. The form changes; the pattern does NOT.

Dubai's real estate market sits at a particularly interesting juncture. Prices have risen ~75% from the 2020 trough. Off-plan speculation accounts for 70% of transactions. UBS ranks Dubai 5th globally for bubble risk. And the city has direct exposure to every signal in this report.

Dubai's modern real estate market only dates to 2002 when freehold ownership opened to foreigners, so we have one full cycle to observe:

The conflict with Iran has introduced a risk that sits outside every analyst forecast issued before March 2026. Drone and missile strikes on the Burj Al Arab, Fairmont The Palm, and Dubai International Airport have shattered the city's most valuable asset: the perception of safety.

Dubai's entire post-2020 boom was built on a narrative - that the city is a neutral, stable, safe haven for global capital fleeing instability elsewhere. Russian oligarchs parked money here after sanctions. Indian HNWIs moved here for tax efficiency. Chinese buyers came for diversification. That narrative is now under military threat.

The transmission mechanism is immediate and does not require a single mortgage default:

1. Capital flight: Foreign buyers account for 58% of Dubai transactions. These are mobile, cash-rich individuals who chose Dubai over Singapore, London, and Miami precisely because of perceived safety. When that perception breaks, capital flows reverse - and 86% cash-buyer status, usually a defense, becomes a vulnerability because cash buyers can exit as quickly as they entered. There are no mortgages locking them in.

2. Tourism and business collapse: Flight suspensions, travel advisories, and evacuations directly hit Dubai's hospitality, retail, and services sectors. Corporate relocations - the engine of rental demand - freeze. Hiring stops. The 6–18 month lag between oil/geopolitical shocks and property prices compresses when the threat is on your doorstep.

3. Insurance and financing withdrawal: War risk premiums surge. International banks and insurers reassess exposure. Construction financing for the 300,000+ unit pipeline becomes harder and more expensive to secure. Developer defaults become plausible - not from leverage, but from project abandonment.

4. Oil price whipsaw: Paradoxically, the Iran conflict could push oil prices sharply higher in the short term (supply disruption, Strait of Hormuz risk), which normally benefits Gulf economies. But elevated oil prices above $90–100/barrel act as a tax on the global economy, triggering recession in import-dependent economies - the same economies whose citizens are buying Dubai property. Our oil research shows a Jupiter-Rahu opposition (Sep 2026–Jun 2027) with an 86% historical hit rate for oil spikes. If oil surges to $100+ during an active regional conflict, the global demand destruction that follows would hit Dubai's RE market on a 12–18 month delay.

Supply tsunami: An estimated 300,000–400,000 units are in the pipeline through 2028. Even at a historical 55% completion rate, that is 175,000+ new units entering a market where prices have already risen 75% from trough. The year 2027 brings the largest single-year delivery in a decade (70,500+ units).

Off-plan dominance: 70–73% of 2025 transactions were off-plan purchases. Off-plan flipping accounts for up to a third of resales. This is speculative activity that reverses quickly when sentiment shifts - and sentiment has already shifted.

UBS Bubble Index: Ranked Dubai 5th globally for bubble risk (score 1.09) - the most significant increase of any city analyzed, measured before the Iran conflict began.

Oil correlation: The 2014 oil crash directly caused a 25–33% decline in Dubai property over six years. The transmission runs through regional liquidity, corporate budgets, hiring, and capital flows. The current environment - oil below $65, OPEC+ adding supply, trade tensions - mirrors the early stages of that episode.

Cash buyers (86%): The absence of leverage means no systemic mortgage cascade. There will be no forced selling by banks. Any correction is driven by sentiment, not margin calls - which makes it slower and potentially shallower than 2008.

Population base: Dubai's population has more than doubled to 4.0 million. Genuine end-user demand from residents provides a floor that didn't exist in 2008 when the city was primarily a speculative playground.

Regulatory improvements: Escrow accounts, DLD oversight, and lower bank RE exposure (~14% of loans vs. ~20% in 2008) reduce the risk of developer fraud and banking contagion.

Diversification: Tourism, tech, financial services, and logistics have broadened the economic base. Dubai is less of a one-trick speculation market than it was in 2008. But diversification does not protect against a war on your border.

The critical difference from every prior Dubai cycle: this is the first time geopolitical risk has directly threatened the physical safety of the city and its residents. In 2008, Dubai's crash was financial - a credit bubble that popped. In 2014, it was structural - oversupply meeting weak oil. In 2026, the risk is existential to the narrative that built the boom: that Dubai is the world's safest place to park capital.

We are not predicting a crash. We are reporting that independent analytical frameworks - quantitative cycle analysis, institutional positioning data, and market structure assessment - converge on the same 18-month window beginning in 2026.

The 18-year cycle does not cause crashes any more than a clock causes midnight. But when the cycle, the affordability data, Buffett's positioning, and the supply pipeline all point to the same window, the prudent response is the one Harrison and Anderson recommend: reduce leverage, build liquidity, and prepare to act when others cannot.

This newsletter is for informational and research purposes only. It does not constitute investment advice.