We investigate, using a unique firm level dataset of nearly 20 million firms in 80 countries, whether differences in the allocation of resources across heterogeneous plants are a significant determinant of cross-country differences in income per worker. Using a monopolistic competitive firm framework to derive our benchmark calibration, we find that the model over-explains income variance. We further explore whether the results are driven by sample biases, calibration assumptions, or modeling choice. We find the same results prevail even in sub-samples in which the data are more reliable, and when we vary the calibration assumptions. This suggests the need for more complex modeling structures. Despite these acknowledged shortcomings, our results suggest that misallocation of resources is a crucial determinant of income dispersion.
Harvard Business School Working Paper 07-086, April 2007