Comment

Abstract

The idea that an important source of cross‐country differences in income is related to differences in the degree ofmisallocation across countries has core appeal to economists. After all, economic efficiency is all about the nature and extent towhich resources are allocated to their highest‐valued use. The basic premise in the paper by Alfaro, Charlton, and Kanczuk is that in well‐functioning economies such as the United States the size distribution of activity largely reflects an efficient allocation of resources. For the core models of the size distribution of activity in the literature, the key implication is that in efficient economies, firms (and establishments) are large because they are the most productive. However, in low income per capita countries, the working conjecture in this paper is that the allocation of resources across firms is distorted. Specifically, the authors explore the implications of recent models that idiosyncratic distortions to the scale of activity in a country will distort the size‐productivity relationship. In this respect, this paper fits into a growing literature seeking to understand the extent of such misallocation. While I am very sympathetic to this line of argument, I have a number of concerns about the identification approach used in this paper. The concerns reflect both conceptual issues and related concerns on whether the data used are sufficient for this identification. Before I proceed to my concerns, it is useful to emphasize the various facets of the analysis that I think are on the right track. For one, there is substantial evidence that there is substantial productivity heterogeneity within industries. The results in Syverson (2004) suggest that the interquartile range for measured revenue‐based total factor productivity within narrowly defined sectors is around 30 log points. In addition, the results in the literature show that there is considerable dispersion and skewness in the size distribution of activity within sectors. These two basic facts offer considerable scope for misallocation to play a role.