Endogenous Local Labour Markets, Regional Aggregation and Agglomeration Economies
Jordy Meekes and Wolter H. J. Hassink.
Local labour markets (LLMs) have received much attention from researchers and policy makers, as they reveal regional differences in economic outcomes such as employment opportunities, wages and housing prices. In research on regional differences in economic outcomes, the set of LLMs within a country is operationalised by using an administrative regional classification, generally defined based on journey-to-work statistics. Examples are the Australian Statistical Areas Level 4 (SA4) or Local Government Area (LGA), or US commuting zone (CZ).
The main motivation of the present paper is that, although most research uses administrative regional classifications to study regional differences, the spatial unit size of an LLM is likely to differ among subgroups of the population. Moreover, we assess the role of the spatial unit size in measuring agglomeration externalities. Agglomeration externalities are derived from the spatial concentration of economic activity through sectoral specialisation and diversity, improving (i) matching of employers to workers and other inputs, (ii) sharing of resources and risk among firms; and (iii) learning through knowledge accumulation by workers. Our research provides a deeper understanding of the size of workers’ LLM and agglomeration externalities, which from a policy perspective is relevant for multiple socio-economic reasons.
First, we define LLMs based on the worker’s commuting outcomes, gender and educational attainment, and show that low-educated workers and female workers are characterised by a relatively small LLM. This finding suggests that place-based policies targeted at workers who are characterised by a relatively small LLM, compared to policies directed at other subgroups of workers, may be more effective if they are specific, local and decentralised.
Second, we find that the positive effect of employment density on workers’ wages, i.e. the urban wage premium, increases when using larger spatial unit sizes to operationalise LLMs, explained by capturing agglomeration externalities that take place at a large spatial scale. This finding suggests urban and regional policies to increase agglomeration benefits and regional productivity growth should tend to be generic and centralised, such as city-region cooperation and geographical upscaling of economic activities.
Third, we show that workers who lose their job in higher-density LLMs experience positive agglomeration externalities on job matching, with more modest losses in wages and again larger density effects using larger spatial unit sizes. This finding is relevant for labour market policies that aim to increase the matching quality of worker to employer or limit wage inequality following negative employment shocks. By gaining a better understanding of social disadvantage in terms of the size of workers’ LLM as well as the winners of the increasing urbanisation of urban areas, this research touches upon an important societal trade-off between equality and efficiency.
November 22, 2019