We know that a large part (half or more) of the difference in incomes per head between rich and poor countries is due to differences in total factor productivity (TFP)--i.e., the efficiency with which capital, skills, labor, and other inputs are transformed into final output--with the rest being due to differences in physical or human capital levels. But what accounts for these TFP differences? A new paper by Chang-Tai Hsieh and Peter Klenow takes us some way into understanding the issues:
We use plant-level data from the Chinese Industrial Microdata (1998-2005), the Indian Annual Survey of Industries (1987-1994) and the U.S. Census of Manufacturing (1977, 1987, 1997) to measure dispersion in the marginal products of capital and labor within individual 4-digit manufacturing sectors in each country. We then measure how much aggregate manufacturing output in China and India would increase if capital and labor were to be reallocated to equalize marginal products across plants within each 4-digit sector to the extent observed in the U.S. The U.S. is a critical benchmark for us, as there may be measurement error and factors omitted from the model (such as adjustment costs and markup variation) that generate gaps in marginal products even in a comparatively undistorted country such as the U.S. We find that moving to “U.S. efficiency” would increase TFP by 30-45% in China and 40-50% in India. The output gains would be roughly twice as large if capital accumulated in response to aggregate TFP gains.
In other words, dispersion in TFP across plants within industries is an important source of the productivity gap across countries. This is a hopeful message insofar as it suggests that poor countries are able to sustain economic activities that are much higher-productivity than what their income levels would suggest and that a lot of economic catchup involves catching up with the productivity frontier within their economies.
But the interpretation of these large TFP differences across plants within the same 4-digit industries is still up for grabs. Is it that credit constraints or political connections prevent more productive plants from expanding, and therefore block economy-wide convergence? Or is what we are observing an inherent sign of economic progress, with resources dynamically being reallocated from low- to high-productivity activities, and in which case these gaps are necessary to sustain growth in transition? UPDATE: Link to paper is now fixed.