Cycle guide · Land and credit
The 18.6-Year Real Estate Cycle
What the 18.6-year cycle is
The 18.6-year cycle is a long-wave idea about land, credit and property. It says that real estate markets, and the economies built on them, move through a repeating pattern of recovery, expansion, a late speculative blow-off and a collapse, and that the full loop takes about 18 to 19 years. Researchers working from the nineteenth century onward noticed that major US downturns came at intervals that clustered around that length, and that many followed periods of land and property speculation.
It is a framework, not a law. It does not tell you what any single price will do next week, and the best evidence for it is historical pattern-matching on a small number of observations. We publish it because it is central to how KriekTrade thinks about the long-term backdrop, and because it is a useful lens even for people who doubt the precise timing. Where we are in the cycle is shown, with its date, on our daily macro page.
The rest of this page sets out where the idea comes from, the record the cycle rests on, its five phases, how it is used in KriekTrade, and its limits.
Where the idea comes from
The cycle is not our invention. A line of economists and analysts built it up over about 150 years:
| Researcher | Date | Role | Contribution |
|---|---|---|---|
| Henry George | 1879 | Economist and reformer | Progress and Poverty laid the theoretical foundation in land-value economics. |
| Homer Hoyt | 1933 | American economist | One Hundred Years of Land Values in Chicago documented recurring land-value cycles of roughly 18 years. |
| Roy Wenzlick | 1930s | Independent researcher | Independent work on real estate cycles in the same period. |
| Fred Foldvary | 1997 | Economist | Wrote in the 1990s that the cycle pointed to a downturn around 2007–2008. |
| Phil Anderson | 1990s onward | Cycle analyst | Refined the cycle length to about 18.6 years. |
| Fred Harrison | 2005 | Economic journalist | Boom Bust (2005) set out the case for a major downturn at the end of the decade. |
Two points are worth keeping in view. First, these researchers worked from different data and in different periods, which is part of why the pattern is taken seriously by those who follow it. Second, several of the forecasts attributed to the framework were made or refined with hindsight, and the 18.6-year figure itself is an estimate from a short run of cycles. Treat the list as the history of an idea, not a performance record.
The historical record
The table lists major US crises that followed periods of land, property or credit speculation, and the number of years between them. The dates and events are standard economic history; the National Bureau of Economic Research (NBER) dates US business cycles from 1854, and the later events below broadly line up with its recorded contractions.
| Year | Event | Gap since prior (years) | Associated speculation |
|---|---|---|---|
| 1819 | Panic of 1819 | — | First major US economic contraction after the early republic’s land and credit expansion |
| 1837 | Panic of 1837 | 18 | Land speculation bubble |
| 1854 | Panic of 1854 | 17 | Railroad speculation |
| 1873 | Panic of 1873 | 19 | Railroad bond speculation |
| 1893 | Panic of 1893 | 20 | Railroad overbuilding |
| 1907 | Panic of 1907 | 14 | Stock market and banking crash |
| 1929 | Great Depression | 22 | Land boom (Florida) and stock speculation |
| 1973 | Oil crisis recession | 44* | Gap includes wartime disruption of the land market* |
| 1989 | Savings & loan crisis | 16 | Commercial real estate bubble |
| 2008 | Global financial crisis | 19 | Subprime mortgage crisis |
| ~2027 | Next cycle peak (framework estimate) | 19 | Cycle completion; not a certainty |
* The 1929 to 1973 gap includes the disruption of the Second World War. Intervals in the table run from 14 to 22 years, apart from that gap, so the cycle is a tendency, not a metronome. Event names and years are well-documented history; the “associated speculation” column follows the cycle researchers’ interpretation.
Looking at the spread of gaps, the average is close to 18 or 19 years, but individual gaps range from 14 to 22. That variation is why we describe the cycle as a window and not a date.
The five phases
We divide the current cycle, counted from the March 2009 market low, into five phases. The years are approximate and are KriekTrade’s own labelling.
| Phase | Years | What it describes |
|---|---|---|
| Recovery (first half) | 2009–2016 | Credit and prices recover from the crisis low; confidence rebuilds slowly. |
| Mid-cycle slowdown | 2016–2020 | A pause or shallow dip in the middle of the expansion. |
| Expansion (second half) | 2020–2025 | Credit growth and land values accelerate. |
| Winner’s Curse | 2025–2026 | A late fake-out pullback that shakes out bulls before the final stretch. This is where our framework places the market as of October 2026. |
| Peak and collapse | 2027–2030 | The most exuberant stretch of the cycle, then the major recession that has followed every prior peak. |
The fifth row is the framework’s expectation, not our forecast of what will happen. The curve below draws the final leg as the framework describes it.
Counting 18.6 years from the March 2009 low gives late 2027. If the true interval is as short as 14 years or as long as 22, the turning point would be anywhere between 2023 and 2031. This is why we say “around 2027” and hold the date loosely.
How we test whether the cycle has ended
Our working view is that the cycle turns when credit stops expanding, not when it merely slows. We look at four questions:
- Is credit still expanding? Lending standards may tighten without lending seizing up.
- Are land and property values still rising? In this framework the peak is marked by land, not by equities.
- Does the late blow-off pattern fit? A final run-up with deteriorating fundamentals is the pattern in earlier cycles.
- Does the timing fit? The arithmetic from the 2009 low points to around 2027, consistent with earlier intervals.
As of October 2026 our assessment is that the cycle has not yet ended and no recession has been declared. That is a statement of the framework’s status, not a prediction, and it will change when the evidence changes. Official recession dating is done by the NBER, usually long after the fact.
How KriekTrade uses the cycle phase
The phase is one of three inputs, with the market risk gauge and the Fed stance, that decide which trade setups KriekTrade publishes. It is set deliberately by Tian Kriek, not computed from a formula, and it takes one of five values: early bull, mid bull, late bull, recession and post-recession. The phase works as a gate on direction. A setup that goes against the direction the cycle allows is held back, whatever the chart looks like.
The gate sits alongside the daily trend rule. Buys are only shown when price has closed above its daily 45-day exponential moving average and the cycle allows buying; sells only when price has closed below it and the cycle allows selling. The present gate setting, with its date, is on the macro page, and you can see the full process on KriekTrade.
The cycle sets the backdrop. It does not predict individual trades, it does not time entries, and it is always combined with technical and fundamental analysis.
What the framework cannot do
- It rests on few observations. About ten crises over two centuries is a small sample, and intervals from 14 to 22 years are too loose to time anything precisely.
- Causes differ. The panics in the table had different triggers: railroads, banking, war, oil, mortgages. Fitting them into one cycle involves judgement.
- Hindsight matters. It is easier to find the pattern in a record that is already written, and some forecasts attributed to the framework were refined after the fact.
- Modern policy changes the game. Central bank intervention, deposit insurance and global capital flows did not exist in the same form in the nineteenth century.
- Markets can stay irrational longer than expected. A bullish bias in the framework is not a promise of gains, and a bearish phase is not a promise of losses.
Use the cycle as one lens, next to the Fed, the yield curve and market risk appetite. Our guides on risk-on and risk-off, yield curve inversion and the Fed and forex cover those.
Key points
- The 18.6-year property cycle is a framework built on over 200 years of crisis records.
- KriekTrade places the market in the Winner’s Curse phase as of October 2026, with the framework’s peak window around 2027.
- Intervals vary from 14 to 22 years, so the date is a window.
- Until the framework says the cycle has ended, the KriekTrade cycle gate keeps allowing buys, always subject to the daily trend rule and to your own risk management.
- None of this is advice, and trading carries a high risk of loss.
Sources and further reading
- National Bureau of Economic Research, US Business Cycle Expansions and Contractions (official dating of US recessions).
- Henry George, Progress and Poverty (1879).
- Homer Hoyt, One Hundred Years of Land Values in Chicago (University of Chicago Press, 1933).
- Fred Harrison, Boom Bust: House Prices, Banking and the Depression of 2010 (Shepheard-Walwyn, 2005).
- Phil Anderson, The Secret Life of Real Estate and Banking (2006).
- Fred Foldvary, The Depression of 2008 (Gamma Foundation, 1997 and later editions).
- Federal Reserve Board, FOMC calendars (context for the policy backdrop).
Charts and diagrams on this page are our own illustrations. Data and facts are cited to the publishing agency; wording and analysis are ours. See the editorial policy.
Risk warning. Trading forex, CFDs, futures and crypto carries a high risk of losing money rapidly, and most retail accounts lose money. For education only; this is not advice. This article is general education written by Tian Kriek of Kriek Institutional Research, the publisher of KriekTrade. It is not personal advice or a recommendation. See the risk warning and disclaimer.