Land Revenue, Not Property Tax, Shapes Which Shenzhen Homes Get Good Schools

In Chinese cities, the quality of a child’s school depends heavily on where the family lives, much as it does in the United States or Europe. But China has no property tax, so the familiar Western explanation — that school quality raises home values, which in turn fund schools through local taxes — cannot apply. A new study of Shenzhen argues that a different fiscal engine drives the same pattern: one-time land-lease payments that local governments collect when they sell development rights, and that these revenues flow disproportionately toward the housing categories that generate them.

The study, published in Frontiers of Urban and Rural Planning, examines how access to public and private primary and middle schools varies across Shenzhen’s three main housing categories: commodity housing, the market-oriented residences built on land sold through competitive bidding; urban villages, informal settlements on collectively owned rural land; and public housing, government-supported affordable units. The authors, Jiawen Yang, Yuling Yang, Yiting Lin and Jiuju Huang, combine a conceptual analysis of China’s land-finance system with a spatial analysis of schools and housing across the city.

The institutional contrast with Western systems is central to the argument. In the United States and many European countries, recurrent property taxes tie school funding continuously to property values within locally governed school districts, creating a self-reinforcing loop in which better schools raise tax bases that pay for better schools. China’s local governments instead capture land value through lump-sum conveyance fees at the moment land-use rights are transferred. Nationwide, land-lease revenues have reached roughly 7 trillion RMB annually over the past five years, and an estimated 40 to 50 percent of that money goes toward public infrastructure, including education. Because the value is captured upfront rather than annually, the authors argue, local governments have strong incentives to invest in amenities that boost land prices before parcels are sold — and weak incentives to maintain services in areas that never generated revenue at all.

Shenzhen offers an instructive setting. The city grew from an urbanized area of three square kilometers in the 1970s to 1,997 square kilometers today, fueled by development-based revenue. Its housing stock is unusually diverse: commodity housing makes up about 17 percent of units, urban villages 51 percent, and public housing 5 percent, with the remaining 27 percent — dormitories and manufacturing compounds — excluded from the analysis. The city also faces acute educational pressure from a large migrant population, many of whom lack local household registration and depend heavily on local schools.

The empirical work draws on a 2016 land-use map, a 2018 building census, 2020 public housing records and residential data from a real estate website, matched against official school district boundaries and school information for 674 primary and middle schools operating at the end of 2019 — 416 public and 258 private. Because exact counts of school-aged children per district were unavailable, the authors estimated demand by applying the city-wide age distribution to each district’s resident population. Public-school enrollment opportunity was measured as a supply-pressure ratio of enrolled students to the resident school-age population, classified as sufficient above 1, mildly short between 0.5 and 1, and severely short below 0.5. Private schools, which operate outside catchment zones, were assessed by distance thresholds of 500 and 1,000 meters for primary schools and 1,000 and 2,000 meters for middle schools. Differences among housing categories were tested with the non-parametric Kruskal-Wallis test.

The results track land-revenue contributions closely. Commodity housing and public housing enjoyed relatively sufficient public primary school provision, with supply-demand ratios between 0.80 and 1.00, while urban villages averaged 0.53 — the most severe shortage. Roughly 56 percent of urban villages faced severe shortages of primary enrollment opportunities, against only about 27 percent of commodity housing. For public middle schools, commodity and public housing averaged ratios above 1, while urban villages averaged 0.69, with about 48 percent experiencing severe shortages. School quality showed the same gradient: about 23 percent of commodity housing sat near prestigious public primary schools, compared with 12.36 percent of urban villages and 9.07 percent of public housing; for prestigious middle schools the figures were about 16 percent, 5.2 percent and 11.78 percent respectively. All differences were statistically significant at the 0.01 level.

Private schools added a stratified second layer. Ordinary private schools clustered around urban villages, filling the quantitative gap left by public under-provision, while prestigious private schools clustered around high-priced commodity housing, where residents can afford premium tuition. Urban villages had the highest overall presence of private schools nearby, public housing the lowest. The authors note that prestigious private schools are often introduced early in housing developments as amenities that lift land values and drive property sales, whereas ordinary private schools emerge as market responses to unmet demand among migrant populations — a dual structure that both compensates for and reinforces existing disparities.

Public housing complicates the revenue story in a revealing way. Although it generates minimal land revenue because its land is administratively allocated rather than sold, talent housing — the subcategory designed to attract skilled professionals — enjoyed educational resources approaching those of commodity housing: about 54 percent of talent housing units were in school districts with sufficient middle school supply, and 17.39 percent sat near prestigious public primary schools, figures comparable to or better than low-medium-priced commodity housing. Public rental housing and economically comfortable housing fared markedly worse, with 8 to 12 percent near prestigious primary schools. The authors attribute talent housing’s advantage to Shenzhen’s economic strategy of using it as a policy instrument to retain skilled labor, either by allocating better resources to it or by siting it near existing high-quality schools.

The authors are careful about the limits of their evidence. The supply-demand indicator captures potential pressure on public-school provision rather than actual enrollment patterns, and the study relies on cross-sectional data for 2019. Educational inequality is also shaped by broader fiscal, institutional and demographic factors, and the lack of parcel-level data on land revenue and its direct allocation to education makes it impossible to precisely quantify the mechanism’s magnitude. The findings for Shenzhen, a city with an unusually high migrant share and a long history of land-finance-driven growth, are presented as illustrative of institutional logics operating across Chinese cities rather than as proof of a universal effect.

The policy implications the authors draw are correspondingly structural. They discuss a gradual, pilot-based introduction of property taxation as a way to build a recurring fiscal base for public services and to motivate governments to invest in long-term operation and maintenance rather than initial construction alone. They also propose incorporating public service indicators into the cadre evaluation system, which currently prioritizes GDP growth. Without reform, they warn, the one-time nature of land revenue could leave even today’s favored commodity housing neighborhoods vulnerable once land development slows — a concern that grows more pressing as China’s property market cools and land-transfer revenues decline.

Subject of Research: How housing-based land-development revenue shapes educational resource allocation across housing categories in Shenzhen, China.

Article Title: The nexus between housing categories and educational inequality in Shenzhen, China: exploring the role of land-development revenue in public service provision

Article References: Yang, J., Yang, Y., Lin, Y., & Huang, J. (2026). The nexus between housing categories and educational inequality in Shenzhen, China: exploring the role of land-development revenue in public service provision. Frontiers of Urban and Rural Planning, 4(1), Article 15. https://doi.org/10.1007/s44243-026-00088-3

Image Credits: AI Generated

DOI: 10.1007/s44243-026-00088-3

Keywords: Shenzhen, educational inequality, land finance, housing categories, commodity housing, urban villages, public housing, school quality, China, public service provision, land-lease revenue, private schools