Distortions, Efficiency and the Size Distribution of Firms

Microdata information about firms' input choices and effective tax liabilities is used to quantify the extent of resource misallocation and efficiency losses due to large tax distortions and limited access to credit. We develop an equilibrium model of firms' behavior in which the tax and credit environments act as a selection mechanism restricting the growth of all but the most productive firms. We show that such a model, parameterized and validated using a variety of data restrictions, has the potential to rationalize several puzzling observations about firms' input choices, size and growth patterns. Counterfactual experiments are designed to gauge the losses associated to different deviations from first-best. We find that firms' optimal responses to the tax distortions are quite effective in reducing efficiency losses. As a consequence, tax distortions only account for 5% of the gap between an undistorted economy and the benchmark. On the other hand limited and expensive access to credit is associated to more significant misallocation of productive resources and leads to larger aggregate efficiency losses of the order of 95% of the gap between an undistorted economy and the benchmark. Our findings highlight the non-negligible quantitative importance of two relatively common distortions in developing economies, and identifies simple mechanisms which might contribute to their low measured TFP.