The Role of the Structural Transformation in Aggregate Productivity^*

Abstract

We investigate the role of sectoral differences in labor productivity in explaining the process of structural transformation -the secular reallocation of labor across sectors -and the time path of aggregate productivity across countries. Using a simple model of the structural transformation that is calibrated to the growth experience of the United States, we measure sectoral labor productivity differences across countries. Productivity differences between rich and poor countries are large in agriculture and services and smaller in manufacturing. Moreover, over time, productivity gaps have been substantially reduced in agriculture and industry but not nearly as much in services. In the model, these sectoral productivity patterns generate implications that are broadly consistent with the cross-country evidence on the structural transformation, aggregate productivity paths, and relative prices. We show that productivity catch-up in industry explains about 50 percent of the gains in aggregate productivity across countries, while low relative productivity in services and the lack of catch-up explains all the experiences of slowdown, stagnation, and decline observed across countries.Keywords: labor productivity, structural transformation, sectoral productivity, employment, hours, cross-country data. JEL Classification: O1,O4. † We thank the editor and three referees for very useful and detailed comments. We also thank comments and suggestions from

Growth and Structural Transformation

“…For example, education and health care are very different activities than retail trade, in that they both represent an investment and tend to use very different skill intensities for the labor that they employ. The work of Jorgenson and Timmer (2011) and Duarte and Restuccia (2012) is a first step in this direction.…”

“…In a recent paper, Duarte and Restuccia (2010) have investigated the importance of these effects in a sample of 29 countries for the period of 1956-2004. They employed a somewhat simplified version of our benchmark model in which labor is the only factor of production (and production functions are linear in labor).…”

“…The initial productivity levels of all countries relative to the US are inferred from the model by requiring that it match the observed employment shares in the initial period. Inputting the sectoral productivity growth rates from the data, Duarte and Restuccia (2010) then simulate the model and compute the implied series for aggregate labor productivity. Even though their model assumes constant productivity growth rates at the sectoral level of each country, it generates large movements in relative aggregate productivity across countries over time.…”

Geography, non-homotheticity, and industrialization: A quantitative analysis

“…Again, using these additional data makes the results from columns 1 and 3 stronger. 13 In Table 2, we gradually build up our results to the full speci…cation (1). In column 1 we include population size; column 2 uses agricultural productivity as an additional regressor; and column 3 includes both population and agricultural productivity.…”

“…The most in ‡uential approaches focus on di¤erences in the income elasticity of demand across sectors (e.g., Murphy et al (1989b); Kongsamut et al (2001)), sector-biased productivity growth (e.g., Ngai and Pissarides (2007)), or a combination of both (e.g., Caselli and Coleman (2001); Duarte and Restuccia (2010)). Traditionally, these approaches have analyzed closed-economy models.…”