Sectoral Growth, Subbotin Distribution, Scaling, U.S. Manufacturing.

C10, O47, O51.

In this paper, we study the growth rates of 4-digit sectors in U.S. manufacturing. Two measures of size (value of shipments, value added) are considered, for each of the 38 years (1959-1996) of a sample of 458 4-digit sectors, drawn from the NBER Manufacturing Productivity database. Whole sample results are partly in line with firm growth facts: (i) sectoral growth rates are distributed according to heavy-tailed Subbotin distributions, with shape coefficient between 1.0 (Laplace) and 1.5; (ii) the volatility of growth rates is decreasing with respect to size, with a scaling exponent varying over time, but always between -0.20 and -0.10. Preliminary analyses on more homogeneous groups cast doubts on the evidence of scaling, but leave basically unaffected the distributional properties of sectoral growth. These results shed light on the role of inter-firm correlations, market concentration, and positive intersectoral feedbacks as drivers of meso-economic dynamics.