In this paper, we trace the causes of regional industrial development in the nineteenth century L... more In this paper, we trace the causes of regional industrial development in the nineteenth century Low Countries by disentangling the complex relationship between industrialisation, technological progress and human capital formation. We use sectoral differences in the application of technology and human capital as the central elements to explain the rise in employment in the manufacturing sector during the nineteenth century, and our findings suggest a re-interpretation of the deskilling debate. To account for differences among manufacturing sectors, we use population and industrial census data, subdivided according to their present-day manufacturing sector equivalents of the International Standard Industrial Classification (ISIC). Instrumental variable regression analysis revealed that employment in the manufacturing sector was influenced by so-called upper- tail knowledge and not by average educational levels, providing empirical proof of a so-called deskilling industrialisation process. However, we find notable differences between manufacturing sectors. The textiles and clothing sectors show few agglomeration effects and limited use of steam-powered engines, and average education levels cannot adequately explain regional industrialisation. In contrast, the location of the fast- growing and innovative machinery-manufacturing sector was more influenced by technology and the availability of human capital, particularly upper-tail knowledge captured by secondary school attendance rates.
In order to study the location patterns of manufacturing firms, and particularly the tendency for... more In order to study the location patterns of manufacturing firms, and particularly the tendency for industry sectors to cluster relative to overall manufacturing, we develop distance-based tests of localization. In order to treat space as continuous rather than using an arbitrary collection of geographical units, we follow the point-pattern methodology of Duranton and Overman (2005, 2008). We apply these techniques on two datasets of Dutch manufacturing firms in two benchmark years, to explore the differences of co-location over time. On the one hand, we will use a dataset of the larger manufacturing establishments in the Netherlands in 1896 and 2010. On the other hand, we will make use of a standardized factory-level dataset in Hunan province, China. Presenting thus two cross sections from two completely different regions, we aim to provide a first empirical test of the agglomeration theories of Marshall (1890) in his period of research and repeat these tests for the recent period.
In this paper, we trace the causes of regional industrial development in the nineteenth century L... more In this paper, we trace the causes of regional industrial development in the nineteenth century Low Countries by disentangling the complex relationship between industrialisation, technological progress and human capital formation. We use sectoral differences in the application of technology and human capital as the central elements to explain the rise in employment in the manufacturing sector during the nineteenth century, and our findings suggest a re-interpretation of the deskilling debate. To account for differences among manufacturing sectors, we use population and industrial census data, subdivided according to their present-day manufacturing sector equivalents of the International Standard Industrial Classification (ISIC). Instrumental variable regression analysis revealed that employment in the manufacturing sector was influenced by so-called upper- tail knowledge and not by average educational levels, providing empirical proof of a so-called deskilling industrialisation process. However, we find notable differences between manufacturing sectors. The textiles and clothing sectors show few agglomeration effects and limited use of steam-powered engines, and average education levels cannot adequately explain regional industrialisation. In contrast, the location of the fast- growing and innovative machinery-manufacturing sector was more influenced by technology and the availability of human capital, particularly upper-tail knowledge captured by secondary school attendance rates.
In order to study the location patterns of manufacturing firms, and particularly the tendency for... more In order to study the location patterns of manufacturing firms, and particularly the tendency for industry sectors to cluster relative to overall manufacturing, we develop distance-based tests of localization. In order to treat space as continuous rather than using an arbitrary collection of geographical units, we follow the point-pattern methodology of Duranton and Overman (2005, 2008). We apply these techniques on two datasets of Dutch manufacturing firms in two benchmark years, to explore the differences of co-location over time. On the one hand, we will use a dataset of the larger manufacturing establishments in the Netherlands in 1896 and 2010. On the other hand, we will make use of a standardized factory-level dataset in Hunan province, China. Presenting thus two cross sections from two completely different regions, we aim to provide a first empirical test of the agglomeration theories of Marshall (1890) in his period of research and repeat these tests for the recent period.
Uploads
Papers
Conference Presentations