Large cities play a disproportionate role in global economic productivity and innovation. Their productivity advantage rests on specialization and the concentration of many diverse skills and capabilities in one place, multiplying economic opportunities and efficiencies. Even so, cities do not appear to develop along one predictable trajectory. Instead, they follow distinct paths reflecting local natural resources, geographic advantages, key strategic decisions and other historical contingencies. The study of cities generally takes this as an established fact – the historical trajectories of cities diverge, rather than converge along some common pattern.
Oddly, however, these distinct pathways of development co-exist with strikingly general patterns which hold across all cities, including quantitative laws for urban scaling and for the distribution and hierarchical structure of businesses. These empirical laws depend primarily on city size – for example, small cities heavily rely on manufacturing-based labour, while large cities instead rely more on cognitive labour.
In a new study, LML External Fellow Hyejin Youn and colleagues document an empirical link between these two contrasting perspectives, and identify a unifying trend in the economic development of cities. They analysed economic activities in 350 U.S. cities between 1998 and 2013, and show that city size determines the typical characteristic industries. In particular, larger and smaller cities show distinct behaviour in the relationship between total employment, Y, and population N. In small cities, Y grows as Nβ, with β < 1, implying that the fraction of population employed decreases with increasing city size. Such cities are characterized by industries such as agriculture and mining. In contrast, large cities follow a scaling behaviour with β > 1, implying an increase in employment fraction with city size. This tends to be the case for industries such as management and professional services. The work shows that cities appear to pass through a universal transition point: a critical population of 1.2 million marks when a city makes a sharp transition in the innovative character its economy. The authors hope their work will improve predictive models for growing cities, and thus help urban policy makers to assess developmental opportunities.
A pre-print of the paper is available at https://arxiv.org/pdf/1810.08330.pdf