innovation, firm growth, Chile, quantile regression, quantile treatment effects

C14, C21, C22, D22, O31

We employ a balanced panel dataset representative of the entire Chilean productive structure in order to investigate the relation between the introduction of innovation and subsequent firm growth in terms of sales. Recent contributions examining the returns to innovation on firm performance have stressed the need of going beyond the analysis of the `average effect for the average firm'. However, previous studies in the case of Latin American economies have often overlooked the importance of analyzing which firms benefit more from the introduction of innovations. Our analysis consists of a series of parametric and non-parametric exercises which take into account the properties of the firm growth distribution. In particular, we adopt quantile treatment effects (QTE) which allow to estimate the effect of the introduction of innovation by comparing firms with a similar propensity to innovate for different quantiles of the firm growth distribution. On one hand, our results indicate that process innovation shows a positive and significant relation with firm growth for those firms located at the 75th and 90th percentiles. On the other, product innovation shows a negative association only for high-growth firms.