The Interrelation between Manufacturing Productivity, Maximum Sectoral Employment and National Income Per Capita

According to the three-sector hypothesis, the national share of employees in the industrial sector will be shrinking after having reached a specific peak level. In line with societal development over time, the tipping point at which deindustrialization starts should be related to a certain income per capita. Predictions from literature showed an inverted-U relationship of manufacturing employment (%) over income per capita (log) for mature countries. By regression analysis, a universal tipping point at which deindustrialization starts was calculated. In this study, this theory was tested on a sample of 12 mature (high income, tipping by 1980) and 25 emerging (upper-middle income, later tipping) countries. No single standard function for all national economies was found, but certain paths related to the general state of economic development of national economies. The tipping point is moving over time, driven by increasing sectoral productivity. Productivity rises result in a shift over time towards higher income (x-axis) and lower relative employment (y-axis). In accordance with existing theory, the country-specific maximum of relative employment in manufacturing is reached at a certain threshold productivity, which again corresponds to a specific national income per capita. As a stylized fact, two falling linear functions of maximum manufacturing employment (%) over GDP per capita (log) were identified for mature and emerging countries. Their divide follows the international division of labour between highly-productive technology owners (mature countries) and less productive sub-suppliers (emerging countries). In addition, the critical manufacturing productivity at which maximum manufacturing employment is reached was analysed as a function of time (tipping year). Corresponding to the findings on the tipping point, the critical manufacturing productivity is rising over time. Two markedly separated rising linear functions for mature and emerging countries were identified.