Heterogeneous effects of e-commerce over technical efficiency in developing countries: Stochastic Frontier Analysis Approach

In developing countries, the business structure usually presents a non-homogeneous environment, between a few large companies and a long tail of unproductive small firms. In Peru, more than 90% of firms are microenterprises, but contribute with less than a third of the private national value added. Moreover, more than 50% of Peruvian enterprises are concentrated in the manufacturing and commerce sector. In this context, despite the positive effects of e-commerce on enterprise efficiency, widely studied in the industrial organization and management literature, only a reduced number of studies delves into the heterogeneity of these effects, particularly according to the firm's size, which is relevant in a polarized business environment. Therefore, the purpose of this study is to evaluate the effect of e-commerce (e-purchases and e-sales) over Peruvian enterprise technical efficiency for the manufacturing industry and wholesale and retail trade sector, considering effects over business size by using a ML Random-Effects Time-Varying Inefficiency Stochastic Frontier model with firm-level survey data collected by the Peruvian National Enterprise Survey (ENE). Our main findings indicate a positive relationship between the fact of carry out e-commerce and enterprise efficiency, but differentiated by business size and economic sector. Whereas e-purchases and e-sales affect positively the efficiency of the entire manufacturing industry, only e-sales reveal a significant effect over wholesale and retail trade sector. Moreover, the effect of e-purchases in the manufacturing industry gets positive and significantly intensify depending on the business size, as well as of e-sales in the wholesale and retail trade sector, revealing a heterogeneous effect of e-commerce over efficiency depending on business size.