Predicting gazelles in Costa Rica: Characteristics and factors of high-growth firms

Friday, March 13, 2015: 10:20 AM
Federico Torres-Carballo, Ph.D. , Business School, Costa Rica Institute of Technology, CAATEC, Cartago, Costa Rica
Ricardo Monge-Gonzalez, Ph.D. , Costa Rica Institute of Technology, CAATEC, Zapote, Costa Rica
We study the topic of high growth rates of firms, to characterize and analyze performance of those firms with a potential for extremely fast growth, generally referred to as “Gazelles” in the literature. We addressed questions like what is the contribution of gazelles to job creation in the country; can we predict the emergence of gazelles in Costa Rica; how heterogeneous are gazelles in this country. In doing so, we follow the work by Davis, Haltiwanger and Schuh (1996) to measure growth of firms and we consider gazelles as defined by the OECD (2009). Access to valuable data from Costa Rica´s Social Security System, the Export Promotion Agency (PROCOMER) and the Ministry of Finance of Costa Rica, has made it possible to increase the theoretical and conceptual understanding of the role of public policies in fostering entrepreneurship in the country. We worked with panel data consists of sixteen thousand firms from 2001 to 2012 selected by systematic random sampling of firms registered in Costa Rica´s Social Security System (CCSS). In this sample, we identified more than 160 gazelles firms from nine hundred firms with 10 employees or more, as definition applied. We found gazelles in Costa Rica are not an industry-specific phenomenon like Henrekson and Johansson (2009) concluded in their meta-analysis. Besides, it seems that there is not a geographic concentration of gazelles in Costa Rica, which rules out the hypothesis about a possible concentration of such enterprises around certain clusters within the country. An important result from the analysis of gazelles is that they are outstanding in employment generation like previous studies remarked. This research is the first that address this topic with panel data in Costa Rica; therefore, we propose a preliminary model to predict emergence of gazelles in Costa Rica. We estimate a probit equation where the dependent variable is the probability of becoming a gazelle. Co-variates are a vector of characteristics of firms and, as a contribution to the economics of entrepreneurship; we explore their experience with exporting and with backward linkages with multinationals.