Spatial growth of employment in "young" services sector mirroring that of early 20th century manufacturing

Klaus Desmet, Esteban Rossi-Hansberg, Mon, 08/13/2007



Manufacturing employment in the US has been deconcentrating for the last 20 to 30 years - the smaller the US county, in terms of manufacturing employment, the greater the increase in employment within the sector. Employment in the services sector, however, differs from this pattern in that within counties with an intermediate level of employment, there has been a greater increase than in the smaller counties (creating an S-shape pattern). The authors of CEPR DP6421 suggest that this difference in evolution may be due to the "age" of the industries concerned.

Measuring the age of an industry as the time elapsed since the last fundamental change in technology, or new "general purpose technology (GPT)", the authors explain that older industries that have already exploited all of the benefits from agglomeration eventually begin to disperse to areas of lower land rents and wages, leading to deconcentration in employment. Younger industries that are still benefiting from knowledge spillovers find it beneficial to agglomerate - hence the fast rate of employment growth in intermediate counties.

The authors find that this theory is consistent with the experience in the US where manufacturing experienced its last GPT - electricity - at the turn of the 20th century, and became a mature industry by the middle of the century once the technology had been implemented everywhere. This led to the deconcentration described above. The last GPT in the services sector - ICT - is a fairly new development, making the service sector a young industry and still able to benefit from agglomeration. The authors find similar patterns in manufacturing and services in Europe.

Journalists are entitled to free DP downloads on request; please contact [email protected]. To learn more about subscribing to CEPR's Discussion Paper Series, please visit the CEPR website.


Topics:  Labour markets


CEPR Policy Research