As centres of human population and civilisation, cities have pushed the boundaries of human productivity and innovation. An outstanding question is whether cities all develop and evolve according to their own particular trajectories, of if instead there are some universal pathways of urban development. Identifying such universal trends in urban growth would help guide policy makers to make decisions to suit the economic and industrial needs of their growing cities.
Recent studies reveal several regular patterns associating urban characteristics with population size. In particular, many different quantities Y change with city population N according to Y(N) ≈ Y0Nβ, suggesting that population size influences many urban properties in a similar way in all cities, although precise estimates of β can be difficult. This scaling pattern also describes economic properties, and reveals quantitative differences between distinct industries. For example, cognitive labour–based industries scale super-linearly with population size (β > 1), while manual labour–based industries instead exhibit sublinear scaling (β < 1). It appears that small cities rely more heavily on manual labour, while large cities employ more cognitive labour.
As LML Fellow Hyejin Youn and colleagues explore in a new paper, however, it remains unclear if scaling relations of this kind apply to a city’s longitudinal evolution over time. To test this, they measure the extent to which individual cities tend to follow the pathway universally prescribed by the scaling law. Using employment data for U.S. cities from 1998 to 2013, they show empirically that the industrial character of cities changes with urban size and industries’ scaling exponent β. Hence, it appears that the longitudinal evolution of urban economies follows a universal process. In particular, the analysis reveals a transition point from a manual labour economy to a cognitive-based innovative economy at a population of around 1.2 million in the United States.
The paper is available at https://advances.sciencemag.org/content/6/34/eaba4934