Within-Country Divides Now Dominate Global Urban Inequality, Study Finds

Global urbanization is often discussed as a contest between rich and poor nations, but a new study argues that the sharpest spatial divides in how cities consume land, concentrate people, and generate wealth now lie inside national borders. Using a globally consistent panel of subnational administrative units spanning 2001 to 2020, researchers found that within-country inequality constitutes the largest component of total spatial inequality across the dimensions they analyzed, while between-country inequality contributes a smaller share. The finding suggests that national averages, the standard currency of global urban assessments, conceal substantial and enduring territorial polarization at the very scale where spatial planning and resource allocation are actually implemented.

The team built a continuous two-decade panel at the second-level administrative-unit scale (ADM2) using the Global Administrative Areas database as its spatial baseline. Three fundamental dimensions of urban expansion were tracked: physical built-up area extracted from a high-resolution annual global urban extent dataset, population density from the WorldPop 1-kilometer global population mosaics, and economic density from downscaled gridded Gross Domestic Product datasets. Crucially, all socioeconomic metrics were confined strictly to dynamic urban built-up footprints rather than expansive administrative metropolitan limits, mitigating a spatial dilution effect in which dense urban cores are statistically distorted by surrounding unpopulated land.

To measure inequality, the researchers applied a hybrid Theil index framework, decomposing total spatial inequality into cross-border international and intra-national components. They then computed spatial Gini coefficients among ADM2 units within individual countries to capture micro-level polarization. The analysis was stratified by the United Nations Conference on Trade and Development’s classification of the Global North and Global South, and further divided into fourteen geographic subregions adapted from the United Nations M49 standard. A five-tier settlement hierarchy, from mega urban settlements of five million or more people down to micro-settlements below 200,000, allowed the team to test whether the scale of a nation’s primary settlement co-occurs with elevated inequality.

The decomposition results reveal a consistent structural pattern across the two decades. For physical built-up area, the Global North’s total land-expansion inequality remained highly stable, with the Theil index hovering below 1.0, while the Global South saw a persistent escalation in overall sprawl inequality. In the South, international gaps widened markedly, yet the subnational territorial divide remained the single largest structural contributor. For population density, the picture was striking: cross-country demographic gaps systematically contracted in both camps, but this between-country convergence coincided with an increase in the within-country component, meaning internal regional sorting accounts for an increasing share of demographic inequality.

Economic density told a different story. Total economic density inequality collapsed significantly over the twenty years in both the Global North and South, driven almost entirely by the compression of subnational gaps. Yet despite this robust regional economic catch-up, the remaining subnational variance still represented most of the total economic inequality architecture by 2020, indicating that localized wealth accumulation remains structurally uneven regardless of broader national-level growth. The spatial mapping also showed that advanced industrialization does not guarantee internal equity: highly industrialized countries including the United States, Belgium, and Germany featured prominently among extreme hotspots of intra-national economic disparity.

The study uncovered a marked North–South asymmetry in what inequality is associated with. In the Global North, spatial polarization was primarily associated with physical land expansion, with the relationship strengthening as urban peripheries grew. In the Global South, inequality was more strongly associated with population growth, with the massive outward drift of Southern data points along the population axis indicating that demographic influx co-occurs with escalating demographic sorting, while physical land expansion showed a remarkably weaker association with settlement patterns. The authors emphasize that these are spatiotemporal associations and structural co-occurrences rather than formally identified causal effects.

Perhaps the most provocative finding concerns settlement hierarchies. In the Global North, the traditional pattern held: countries dominated by mega urban settlements of more than five million people exhibited the highest demographic and economic polarization around primary nodes. But in the Global South, a paradox emerged. It was the medium- and small-urban-settlement-driven nations, those whose largest settlements hold fewer than one million people, that traced the highest trajectory of demographic inequality in the developing world. The researchers suggest that in the absence of a mature, globally integrated metropolitan network to absorb mass migration, populations in these nations hyper-concentrate into solitary, under-equipped primary towns, leaving the periphery demographically hollowed and structurally deprived. In the Global South, the economic penalty has also recently shifted, with large-urban-settlement-driven trajectories overtaking mega urban settlements, suggesting that acute economic imbalances currently peak during volatile intermediate phases of rapid urbanization.

Longitudinal trajectory clustering using Dynamic Time Warping classified national inequality curves into descriptive patterns of aggravating or persistent, volatile or stagnant, and converging or improving inequality. The results showed no uniform path toward convergence. For built-up area inequality, a plurality of nations in both the Global North (37.2 percent) and Global South (42.3 percent) fell into the aggravating cluster. Population-density inequality showed marked persistence, with 65.1 percent of Northern countries assigned to the aggravating trajectory compared with 14.0 percent in the convergence cluster, while the South exhibited greater diversity with 33.7 percent converging. Economic density trajectories were bipolar, divided between successful structural corrections (44.2 percent in the North, 41.3 percent in the South) and compounding localized disparities (41.9 percent and 32.7 percent). Separately, a Coupling Coordination Degree model revealed that over 2001–2020 the relationship between population-density and economic-density inequality shifted toward lower coordination, with a more pronounced movement in the Global South, indicating a widening mismatch between demographic concentration and localized economic output.

The analysis also engaged with the United Nations’ SDG 11.3.1 indicator, the ratio of land consumption rate to population growth rate. Downscaled to subnational units, the within-country distribution of this ratio showed a higher median in the Global North but a wider distribution and longer upper tail in the Global South, indicating greater subnational heterogeneity in the land-population growth relationship in developing regions. The authors note that the indicator’s proportional-growth assumption may penalize compact densification, which is why they complemented it with a Tapio decoupling model capturing non-linear relationships between land expansion and population change.

The team is candid about the limitations of its framework. The number and geographic extent of administrative units vary substantially across countries, and country-level Gini coefficients were moderately associated with the number of analyzed units, meaning the reported coefficients describe inequality across the adopted ADM2 delineations rather than scale-invariant measures. Sensitivity analyses using finite-sample corrections, fixed-size sampling, area weighting, and scale-restricted samples generally preserved the positive South–North difference in country-mean inequality, particularly for built-up area and GDP density, but the comparatively small population-density difference was less stable and should be interpreted cautiously. The framework also does not incorporate transport networks, housing stocks, or utility infrastructure, so its references to spatial mismatch are strictly limited to the decoupling of population agglomeration and economic output within built-up footprints. Even so, the core message stands: global urban inequality is a multidimensional territorial process operating within countries, and one-size-fits-all national metrics risk obscuring the localized gradients of deprivation where equitable and sustainable urbanization will ultimately be won or lost.

Subject of Research: Intra-national spatial inequality in urban land, population, and economic activity across the Global North and South

Article Title: Divergent trajectories of global urbanization: Intra-national spatial inequality and structural mismatch across the Global North and South

Article References: Shen, C., Zhang, Y., Wang, Y., & Wu, T. (2026). Divergent trajectories of global urbanization: Intra-national spatial inequality and structural mismatch across the Global North and South. Frontiers of Urban and Rural Planning, 4(1), Article 25. https://doi.org/10.1007/s44243-026-00096-3

Image Credits: AI Generated

DOI: 10.1007/s44243-026-00096-3

Keywords: urbanization, spatial inequality, subnational analysis, Global South, Global North, built-up area, population density, economic density, Theil index, Gini coefficient, SDG 11.3.1, urban planning