Abstract
We derive a measure of the relatedness between economic activities based on weighted correlations of local employment shares, and use this measure to estimate city and activity complexity. Our approach extends discrete measures used in previous studies by recognising the extent of activities' local over-representation and by adjusting for differences in signal quality between geographic areas with different sizes. We examine the contribution of relatedness and complexity to urban employment growth, using 1981–2013 census data from New Zealand. Complex activities experienced faster employment growth during our period of study, especially in complex cities. However, this growth was not significantly stronger in cities more dense with related activities. Relatedness and complexity appear to be most relevant for analysing how large, complex cities grow, and are less informative for understanding employment dynamics in small, less complex cities.
Highlights
The spatial concentration of economic activities in cities generates agglomeration economies arising from labour market pooling, input sharing, and knowledge spillovers (Marshall, 1920)
We examine the contribution of relatedness and complexity to urban employment growth, using 1981–2013 census data from New Zealand
Âa02A\{a} RCAca0 Raa0 Âa02A\{a} Raa0 of relatedness density, which estimates the share of activity a’s relatedness with all other activities that is contributed by locally over-represented activities
Summary
The spatial concentration of economic activities in cities generates agglomeration economies arising from labour market pooling, input sharing, and knowledge spillovers (Marshall, 1920). We examine the contribution of relatedness and complexity to urban employment growth, using 1981–2013 census data from New Zealand These data cover a range of urban areas that are smaller than, but contain similar activities to, previously studied regions. Relatedness and complexity appear to be relevant for analysing how large, complex cities grow, but offer little information about local growth trajectories in small areas This result is consistent with the idea propagated throughout the urban economics literature that cities are dense networks of interacting activities; in our data, the benefits of such interaction are more apparent in larger cities, in which activity networks are more dense and in which people with complementary skills interact more frequently. Appendix A offers a brief primer on networks and graph theory, ideas from which we use throughout the paper
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
