We use the non-parametric Data Envelopment Analysis (DEA) and the parametric Stochastic Frontier Analysis (SFA) to workout efficiency and productivity of the ICT sector across 50 countries for the time period 2016 and 2020. By considering OECD -DSTRI index as bad output, the study intends to estimate the influence of digital trade regulations (or digital services trade restrictiveness) on productivity and efficiency of the ICT sector across countries. The SFA scores are compared with the DEA efficiency scores and Malmquist Index based on four DEA problems are used to work out the total factor productivity of the ICT sector across countries and over two time periods. Cross country regression is run to understand the impact of ICT efficiency and productivity on average annual growth rate of GDP per-capita among other control factors. We use network readiness index of the World Economic Forum along with OECD-DSTRI index as our database.
Broadly, our results are three-fold. Firstly, we find that trade restrictiveness has very little to do with technical efficiency based on SFA frontier analysis. Secondly, we find no statistical difference in the technical efficiency scores of high and middle-income countries based on DEA analysis. We also found evidence of Asian countries faring better than the European Union countries in context of technical efficiencies. Thirdly, from our empirical results it seems that adoption of ICT and the so-called disruptive 4IR technologies have replaced labour and henceforth, we hypothesize that higher unemployment leads to higher growth rate along with a positive impact of network readiness on the growth rates. It seems that net neutrality and addressing pernicious regulations related to the ICT sector - barrier to competition, among others like foreign entry, mobility of people, regulatory transparency across countries can improve the productivity and efficiency of the ICT sector.