One or more growth processes? Evidence from new Italian firms A. Bonaccorsi Æ S. Giannangeli Accepted: 15 March 2008 / Published online: 9 September 2008 Ó Springer Science+Business Media, LLC. 2008 Abstract The paper proposes a descriptive model of firm growth which assumes that learning processes are conditional on existing competencies of the firm and its founders. We suggest the idea that firms who do not grow under all realizations of potentially profitable opportunities can be considered as zero- learning. In order to explain why some firms are zero-learning and how these are different from growing ones, we put forward and discuss the hypothesis that both start-up size and founders’ pre- entry history affect the firm’s ability to adjust to market. Using a sample of 3,905 Italian firms born in 1999 and 2000, we find that individual competencies influence start-up size, but not directly growth. We also find a significant nonlinear relationship between start-up size and growth, implying that firms which were born smaller than a given size grow significantly less. The results support the hypothesis that multiple and heterogeneous growth processes coexist and that the growth process of microfirms below a given organizational threshold may be structurally different, in the long run, from that of firms starting up above that threshold. Keywords Firm growth Á Entrepreneurship Á Firm capabilities JEL Classifications L25 Á L26 1 Introduction Much of the growth-of-firm literature in the 1980s and 1990s has been shaped by a revival of interest in the so-called law of proportionate effect proposed by Gibrat (1931). One of the implications of this law is that the growth rates of firms in a given population are independent of their initial sizes. Many theoretical and econometric advances have been made since the original formulation of Gibrat’s law [see Sutton’s (1997) survey of ‘‘Gibrat’s legacy’’]. One of the streams of research spurred by the studies of Evans (1987a, b) and Dunne et al. (1988, 1989) has focused on the relationship between size and growth of firms. The ‘‘stylized fact’’ emerging from these seminal studies, and confirmed by subsequent empirical research (Hall 1987; Acs and Audretsch 1987; Mata and Portugal 1994; Wagner 1994; Reynolds 1997; Harhoff et al. 1998; Shane 2001; Lotti and Santarelli 2004), is that growth rates of firms, conditional on survival, are not independent of their sizes but tend to be negatively correlated with them. A. Bonaccorsi School of Engineering, University of Pisa, via Diotisalvi, 2, Pisa 56126, Italy S. Giannangeli (&) Laboratory of Economics and Management, Sant’ Anna School of Advanced Studies, Piazza dei Martiri della Liberta `, 33, Pisa 56127, Italy e-mail: s.giannangeli@sssup.it 123 Small Bus Econ (2010) 35:137–152 DOI 10.1007/s11187-008-9131-0