A Count Data Model of Technology Adoption

This paper investigates the main determinants of the adoption of flexible production technologies (FPTs), using a plant-level dataset of Portuguese manufacturing industry. Besides using a new dataset, this paper extends the framework of previous studies on technology adoption by taking into account the effect of demand uncertainty as an additional determinant of adoption. In order to examine this relationship, several econometric models for count data are estimated. These models deal with the discrete nature of the dependent variable and firm specific unobservable characteristics arising from the cross-section context. The main findings of the paper are: (i) rank effects and technological regimes are important determinants of technology adoption, as put forward in previous models of technology diffusion; (ii) demand uncertainty has a significant positive impact on the likelihood of adopting FPTs, which suggests that technological heterogeneity is important when modelling firms' investment decisions; (iii) the estimates were improved after controlling for excess zeros and overdispersion that characterizes our data.