The Potential and Utility of Land Value Taxation: A Theoretical Framework and Simulation for China (Hou, Kumhof & Shao, 2026)

Summary

"The Potential and Utility of Land Value Taxation: A Theoretical Framework and Simulation for China," by Yilin Hou (Syracuse University), Michael Kumhof (CEPR, Centre for Macroeconomics), and Lei Shao (Central University of Finance and Economics, Beijing), appeared as CESifo Working Paper No. 12476. It builds a three-sector DSGE (dynamic stochastic general equilibrium) model — extending Kumhof et al.'s 2021 "Post-Corona Balanced-Budget Super-Stimulus: The Case for Shifting Taxes onto Land" — calibrated to China's 2015 economy (with a full robustness re-run on 2023 data), to compare financing a consumption-tax cut with either higher income taxes or a higher land value tax.

The Simulation

The paper estimates land at 45% of China's total private fixed assets in the baseline calibration (127.87 trillion CNY of an estimated 284.79 trillion CNY in net household assets, following Li et al. 2019 and Herd 2020), with 55% used as an alternative/sensitivity case. The core experiment: cut the gross consumption tax rate from 19.43% to roughly 13%, financed either by raising labor/capital/land income taxes (the PIT regime) or by raising the tax on the capitalized asset value of land from 3.72% to 13.72% — a flat 10-percentage-point increase, phased in linearly over 20 years (the LVT regime).

The results diverge sharply. Under the PIT regime, headline GDP falls by roughly 8% relative to trend (GDP excluding taxes on production falls by ~4.5–5%). Under the LVT regime at the 45% land-share baseline, headline GDP stays essentially flat relative to trend, while GDP excluding taxes on production rises by over 3%. At the 55% land-share alternative, headline GDP rises 5% and adjusted GDP rises roughly 9% — the source of the "3–9%" range in this page's excerpt. Welfare, measured as consumption-equivalent variation, rises 8% of steady-state consumption under the LVT regime (45% land share) versus 6.6% under the PIT regime. Land prices fall roughly 60% under the LVT regime, with real estate prices (land plus structures) falling around 30%. The 2023 robustness check finds qualitatively similar but quantitatively stronger results favoring LVT, because lower steady-state growth in the more recent calibration reduces land price sensitivity to the tax.

Relation to the Georgist Case

This is a rigorous, well-cited engagement with the Georgist tradition specifically, not just a generic tax-shifting exercise. The paper's Section II, "Land Taxation in the History of Economic Thought," discusses Henry George directly, citing Progress and Poverty, and states that land taxation "was highly recommended in the late-19th century by thinkers including Henry George as an efficient and powerful tax for economic development." It also cites Mason Gaffney and Gaffney & Harrison (1994) — including their argument that objections to taxing land rights echo historical defenses of slavery — plus Andelson & Gaffney (1979) and Stiglitz's land-and-inequality work. The result is a concrete, quantified demonstration of the Henry George Theorem's practical logic applied to a specific, large economy where land is an unusually large share of wealth, directly complementing the wiki's existing China land-policy cluster.

Nuances and Limits

• A calibrated DSGE simulation, not a natural experiment or observed policy change. The results describe a model's prediction under specific structural assumptions (nested CES production, Greenwood-Hercowitz-Huffman preferences), not causally-identified real-world evidence.
• China's land tenure is distinctive — land is state-owned and allocated via leasehold, making a "land value tax" here closer to adjusting the terms of existing state land-lease revenue than introducing a wholly new private-property tax, a structural difference from most of the wiki's other LVT simulations.
• Full text read directly (A-claim). The complete 61-page working paper was read, including the model structure, calibration, and results tables.

Bears On

• Concept: Henry George Theorem — a large, quantified simulation of the theorem's core logic (shifting tax burden onto land improves welfare) applied to a specific major economy.
• Research: Xu, Huang & Li: Did Henry George Inspire China's Economic Miracle? — a different (institutional/historical) argument for Georgist logic in China; this paper supplies the missing quantitative macro simulation.
• Research: Gaffney: A Real-Assets Model of Economic Crises (China) — the macro-financial risk side of China's land-dependent growth model this paper's tax-shifting proposal would help address.

See Also

• Land Value Tax
• Henry George Theorem
• Xu, Huang & Li: Did Henry George Inspire China's Economic Miracle?
• Gaffney: A Real-Assets Model of Economic Crises (China)
• Liu: Shanghai and Chongqing's Real Estate Tax Pilots

Sources

1. Yilin Hou, Michael Kumhof & Lei Shao (2026), "The Potential and Utility of Land Value Taxation: A Theoretical Framework and Simulation for China," CESifo Working Paper No.
2. SSRN (blocked to this session) · full text read directly at ifo.de/DocDL/cesifo1_wp12476.pdf 2026-08-29 — used for the model structure, the 45%/55% land-share calibration, the 3.72%→13.72% LVT rate path, the PIT-vs-LVT GDP and welfare comparisons, the land/real-estate price effects, the 2023 robustness check, and the paper's direct engagement with Henry George, Gaffney, and Andelson & Gaffney (A-claim; full text).