Empirical analyses on aggregated datasets have revealed a common exponential behavior in the shape of the probability density of corporate growth rates. We present clear-cut evidence on this topic using disaggregated data. We explain the observed regularities proposing a model in which firms' ability to take up new business opportunities increases with the number of opportunities already exploited. A theoretical result is presented for the limiting case in which the number of firms and opportunities go to infinity. Moreover, using simulations, we show that even in a small industry the agreement with asymptotic results is almost complete. * Scuola Superiore Sant'Anna; bottazzi@sssup.it; secchi@sssup.it. Support by theItalian Ministry of Education, University and Research MIUR (grant no. A.AMCE.E4002GD) and the S. Anna School of Advanced Studies, Pisa (grant no. E6003GB) is gratefully acknowledged. The authors thank M. Anoufriev, C. Castaldi, G. Dosi, S. Modica, and M. Sylos Labini for helpful discussions and A. Pakes and two anonymous referees for insightful comments. The research would not have been possible without the invaluable help of the Italian Statistical Office (ISTAT) and in particular of Roberto Monducci. The usual disclaimers apply.

No abstract

The size distribution and growth rate dynamics of U.S. companieshave been extensively studied by many authors. In this paper,using the COMPUSTAT database, we extend the analysis todisaggregated data, studying 15 sectors of the U.S.manufacturing industry. The sectoral investigation reveals thepresence of general statistical properties that can be consideredvalid across all the studied sectors. In particular, theprobability density of firms growth rates invariably displays acharacteristic tent shape and the relation between the size of afirm and the variance of its rates of growth is characterized, indifferent sectors, by very similar scaling relations. Thepresence of characteristics that are robust and sectoral invarianthints at the existence of generic statistical properties shapingthe dynamic of firms across the whole industry

The short run effects of financial constraints (FCs) on the expected growth rate of firms and their long-term implications on the evolution of the firm size distribution have been recently investigated by several scholars. In this paper we extend the analysis to a wider and largely unexplored range of possible FCs effects, including the autoregressive and heteroskedastic structure of the firm growth process and the degree of asymmetry in the distribution of growth shocks. We measure FCs with an official credit rating index which directly captures the borrowers' opinion on a firm's financial soundness and, consequently, the availability and cost of external resources. Our investigations reveal that FCs operate through several channels. In the short term, FCs reduce the average firm growth rate, induce anti-correlation in growth shocks and reduce the dependence of growth rates volatility on size. Financing constraints also operate through asymmetric threshold effects, both preventing potentially fast growing firms from enjoying attractive growth opportunities, and further deteriorating the growth prospects of already slow growing firms. The sub-diffusive nature of the growth process of constrained firms is compatible with the distinctive properties of their size distribution.

This work explores and compares some basic properties of corporate growth process at both aggregate manufacturing level and disaggregated sectoral levels. Using an extensive dataset on Italian manufacturing firms, we investigate which properties of firm growth dynamics are robust under disaggregation. We compare the results obtained with three different definitions of firm size, namely total sales, number of employees and value added. Our analysis suggests that while different sectors are characterized by significant differences in firm size distributions, in the degrees of concentration and in the autoregressive structure of the growth processes, there are also regularities which hold across all of them, such as the approximate unit root nature of the growth process and the power exponential shape of the growth rates density. Together, these “stylized facts” suggest challenging puzzles on the drivers of corporate growth and the resulting industrial structures. Copyright Springer 2007aggregation, concentration, firm growth, size distribution, C1, D2, L1,