Land and the next financial crisis. A response to Fred Harrison’s “Rent induced Recessions: When risk is not a risk”

25th June 2026 

Key takeaways

The value of land is large and follows cycles with peculiar regularity. If Harrison is right and the land cycle is edging towards its peak at this moment, at a time where so many other vectors are also lined up for vulnerability, such as highly indebted sovereigns, overinflated tech stocks, central banks with policy options reduced by inflationary pressures and banks highly exposed to land, any trigger can create a financial downwards spiral between land, banks and credit. And if land does have that power, Harrison proposes common sensical fiscal solutions that not only reduce deadweight losses from taxes on labour and capital, but also tame the land cycle’s role in deepening financial crises.
 

Land and the next financial crisis

A response to Fred Harrison’s “Rent induced Recessions: When risk is not a risk”

Harrison is right that land is under-researched, under-measured, and under-used in macro-finance and public-finance analysis. I shall mention existing work that already shows that land and collateral matter for credit cycles. If land rents are capitalised into land prices and privately captured, then land taxation has both strong public-finance appeal and important financial-stability implications. What is as yet missing is research on the amplifications and interactions of the land cycle with the financial/credit cycle that lead to the remarkable regularity of the land cycle and the fact that its downturns tend to coincide with and worsen financial crises. If Harrison is right and the land cycle is edging towards its peak at this moment, at a time where so many other vectors are also lined up for vulnerability, such as highly indebted sovereigns, overinflated tech stocks, central banks with policy options reduced by inflationary pressures and banks highly exposed to land (including directly and indirectly through data centre related exposures), it is not hard to imagine that any one of a large number of triggers can create a financial downwards spiral between land, banks and credit. And if land does have that power, Harrison proposes some common sensical fiscal solutions that are not only reduce deadweight losses from taxes on labour and capital, but at the same time tame the land cycle’s role in deepening financial crises.

Neglected land. Very few agencies gather and make land price data accessible, as Harrison writes in his essay, even though land is believed to be the largest source of material wealth. In the UK for instance, the ONS estimates that of the £13.1tn UK net worth in 2024, UK land is worth £7.1tn and the FTSE All-Share £2.74tn. Since researchers require quality data for their work, little is known empirically about the land market across regions and countries. Furthermore, publishable papers in economics increasingly require clean causal identification, whereas macroeconomic questions about land often involve few observations, noisy data, and long cycles. 

What existing research already shows. While the propagation mechanism is not fully understood, from past LSE work we do know a few things of interest that can guide future research. 

Kiyotaki and Moore (KM) (“Credit cycles,” Journal of Political Economy, 1997) model an economy with land, capital and credit and generate endogenous credit cycles, magnified by and magnifying the land price cycle, since land acts as collateral to debt. When calibrating their model, they find a cycle of about 10 years, and a peak in land prices that leads the debt cycle by 7 quarters. In KM, land is the only tradable net asset with cycles (they assume capital is not tradable so does not capitalise into a market price). This simplification may help explain the ten-year cycle in their calibration. Further research may show that when adding other asset classes, including equities and sovereign bonds, together with financial amplifiers like banks, the models may generate a land cycle closer to 19 years. A visual inspection of land and property-price series often suggests a 16-19-year cycle. To verify this frequency, spectral analysis on post 1950 data done by the author also finds that 19 years is by far the strongest frequency, followed by 13 years and then 10 years, consistently across western countries. What is also missing in this paper is the inter-generational effect, introduced in Hirano and Stiglitz (“Land Speculation and Wobbly Dynamics with Endogenous Phase Transitions,” LSE Centre for Macroeconomics discussion paper 2022-01). The latter paper is a OLG model with two-period lived agents (young and old), so not designed to be calibrated to the actual land cycle length, but it does have the pleasing property that in a rational expectations equilibrium, too high land prices must mechanically lead to an inevitable downturn.

Incorporating Harrison's argument. Knowing these things of course makes matters worse in Harrison’s mind, because if we know how important land is in the propagation of cycles, surely policy makers should keep a keen eye on the developments in the land markets. 

While the macro outlook does not look promising, Harrison does not believe the doom narrative that public funding cannot be raised to stem the decline. The surplus that the economy generates is in his view large enough to fund the community investments that are required for a healthy society. The problem is that the current taxation rules create excessive deadweight losses, thus leaving public investment too low and public debt too high, while also failing to tax much of the unearned surplus. He believes this overall taxable but untapped surplus arising from rent seeking, or what Harrison calls “cheating,” is large enough to remove the distortionary taxes on labour and capital.

Cheating is related to Harrison's Law of Absorption, stating that any investment, be it public and financed by the public purse, or private, generates surplus that is captured in the long run to a significant extent by land prices. It follows that land owners privately capture the benefits of investments made by others, including the public, a form of cheating. To my knowledge, no macro finance model with fiscal policies has incorporated that rent generation.

Harrison makes the case for incorporating rent into models forcefully in his paper, and more broadly in his book Cheating (Shepheard-Walwyn, 2026). If the unearned surplus was taxed, he argues, taxes on labour and capital could be reduced, welfare losses could be lowered, and society would be better off. The argument is strengthened by the fact that the supply of land is inelastic, making land rent a relatively efficient tax base. In its simplest form, the rule is: “pay for what you get.” 

Harrison points out that this new fiscal rule has one more immediate benefit for financial stability: the land cycle that is at the critical amplifier of crises is muted because land can no longer capitalise rent. Few economists trained in price theory would disagree with the logic. The difficulty is political: lowering taxes on human and physical capital while raising taxes on rent requires political will.

Studying the transition. Harrison’s proposal calls for more finance research. If land rent is taxed, land (and thus property prices for land is one component) prices fall, and with them the value of the collateral secured by land and property. This has, beyond the financial stability benefits from a muted land cycle, two further consequences that need to be considered. 

First, land as a collateral becomes less available. To the extent, as in Koyotaki and Moore, that collateral is required for productive credit, substitutes for collateral need to be found to guarantee business loans. This problem should not be overstated, however. Lower land values also mean that lower mortgage loans are required to purchase land and property. The main difficulty concerns legacy loans. Given the remaining duration of many such loans, it should be possible to implement Harrison’s proposals gradually, perhaps over a decade, in order to reduce the risk of disruption.

Credit taken out with property as collateral to fund or smooth consumption will also be harder to obtain, though in the UK is unlikely to be large (BofE statistics do not seem to allow me to infer what fraction of loans collateralised by property are taken out for consumption purposes, but I expect that to be low single percents of all mortgage loans). 

The second issue is the fact that land and property are currently a large source of household wealth. Since consumption, both during the working life and in retirement, depends on permanent income and wealth, any transition to taxing rent must take seriously the wealth and redistribution effects arising from lower land prices, and how to redirect household savings into productive investments instead. The value of those productive assets increases since gains can accrue tax free and decisions more efficient as welfare losses are removed. What is left is a problem of redistribution.

Thoughts on land and the next systemic crisis. On the narrower question of financial stability, which is of particular interest to this forum, I find myself in agreement with much of Harrison’s diagnosis. The land cycle is at the least a major contributor to the regularity and severity of financial crises. Kiyotaki and Moore as well as a sequel paper to Hirano and Stiglitz identify one mechanism: land values affect collateral, and collateral affects credit. Work by the SRC on endogenous risk identifies another: banks’ direct and indirect exposure to land - including through property lending, private credit and data centres - creates a powerful transmission and reinforcement mechanism between the land cycle, the financial cycle and the business cycle. When a downturn in the credit cycle coincides with a downturn in the land market, the result can be systemic.

This is further underpinned by Harrison's observation on the relative invisibility of the land cycle from official statistics. If a risk driver is plainly visible and understood, one would expect this driver to stop being a driver simply because policy makers and regulators would use the driver to implement policies that reduce this risk, a version of the Goodhart Law in action. What Harrison says is that even though official land statistics are not prominently visible, policy makers do have sufficient information to ascertain the land cycle. The problem is that they do not fully use that information.

At this juncture in May 2026, the land and property markets have started turning down in a few locations, especially in real terms. As per BIS data (https://data.bis.org/topics/RPP/tables-and-dashboards), continental European countries have recently suffered large, even double digit, real price drops the last couple of years (e.g. Germany, France, Luxembourg) with the expected pressures on banks’ balance sheets. The UK and the US markets have also started to pause and edge lower in real terms, and the US house builder ETF XHB has largely stalled the last couple of years (albeit with quite a bit of volatility).

This forum is therefore an appropriate place to raise awareness of the land cycle, including Harrison’s warning that it may peak this year and turn down thereafter, provided that the historical cycle frequencies observed over the last two centuries have not changed. Major wars can delay cycle turns: indeed the cycles did pause for a while with WWII. It is also an encouragement for academics to pursue further research into what determines the length of the land cycle, so that we can inform policymaking more forcefully.

One promising avenue would be to construct a model of an economy with realistic overlapping generations, given the likely generational aspect to the cycle’s regularity. Another would be to incorporate a financial sector that interacts with the land cycle through leverage and liquidity. Both models should allow the fiscal authorities to optimise their fiscal rules over distortionary taxes as well as taxes on land rent.

In the meantime, we may be wise to follow Harrison's suggestion and prepare our answer for King Charles, should we be blessed with him agreeing to open the new Firoz Lalji Global Hub building at LSE in the next few years.                 

To comment, please go to the discussion forum