Structural transformation, productivity growth and input-output linkages – Ákos Valentinyi
As a country develops, the workforce gets reallocated first from agriculture to manufacturing and services, and later from agriculture and manufacturing to services. This results in an inverted U-shaped relationship between the manufacturing share of employment and GDP per capita. For many poor countries yet to industrialise, this curve is yet to show itself. However, interestingly, in countries where the process started more recently, the peak of the curve has occurred at a lower GDP per capita than in countries where industrialisation started much earlier, the current developed countries. This may suggest that late developers deindustrialise too early and get stuck in middle-income status.
This pathfinding paper will examine two key policy questions: first, how to ignite industrialisation in poor countries where the peak of the curve is yet to be reached; and second, how to avoid the middle-income trap for late developers. To answer these two questions, we need to understand the drivers of sector productivity growth. The standard approach, much used up until now, has ignored intersectoral input-output linkages. New research, largely in higher-income countries countries, has emphasised these linkages in understanding sources of productivity growth and designing policy. Input-output analysis allows identification of key industries which are the primary reasons for slow sector and aggregate productivity growth. A key constraint on this approach in low-income countries has been a lack of data. This paper will address how and why more of these data must be collected and how these more novel methods can be effectively applied in poor countries, especially sub-Saharan Africa.