The Urban Niche

Economic growth is analyzed at the national level because that is where the data is. But cities generate the bulk of global output. The nation is an administrative container, not an economic unit.

arXiv:2603.16007 constructs GDP trajectories for 8,808 functional urban areas across 165 countries over 26 years using satellite nighttime light data, then clusters the full temporal trajectories. The result: 17 distinct, persistent growth regimes that are better described as ecological niches than as positions on a development ladder.

Cities within the same country frequently belong to different regimes. Structurally similar cities on different continents share the same one. Regime membership explains 16% of within-country growth variance beyond country fixed effects — a substantial fraction given that conventional analysis assumes growth is nationally determined.

National-level convergence — the finding that poorer countries grow faster — turns out to be an aggregation artifact. Conditional convergence operates within regimes, not globally. The pattern exists because averaging cities within different regimes into a single national number creates an apparent trend that is not present at the level where growth actually happens.

A directed propagation network reveals that shocks transmit along lines of structural similarity, not geographic proximity. Advanced economies export disturbances; emerging economies absorb or amplify them. The transmission channel is economic structure, not trade linkage or physical adjacency.

Within-country spatial inequality declines with industrialization maturity — growth initially concentrates in leading cities before diffusing across the urban system. The global economy is an ecology of heterogeneous urban growth regimes, not a collection of nations on a shared development path. The unit of analysis was wrong.