Economic and environmental performance of the belt and road countries under convex and nonconvex production technologies

The Belt and Road Initiative (BRI) is expected to boost the economic and technological cooperation among the participating countries. In this paper, we argue that the economic and environmental efficiency can be used as measures identifying the best practice in the BRI. However, the results of the analysis may be affected by perturbations in the data and outlying observations. The outliers may lead to biased policy implications. Unlike previous research, this paper compares economic and environmental performance in developing countries using both convex and nonconvex production technologies based on a nonparametric framework. Specifically, this paper extends the by-production model with nonconvex technology. The model allows reducing the impact of extreme production plans (e.g., small or large countries). The resampling approach is applied for the by-production framework. The research covers 61 countries along the Belt and Road during 2000−2017. Our results indicate that the influence of outliers is more significant for convex technology than nonconvex technology. The use of the robust approach is particularly important when a convex technology is assumed. The proposed approach also allowed identifying BRI countries that are significantly over- or underperforming in either economic or environmental efficiency.