Forecasting Energy Efficiency in Manufacturing: Impact of Technological Progress in Productive Service and Commodity Trades

This paper employs the theory of biased technological progress to assess the effects of technological advancements across diverse trades, with a particular emphasis on predicting energy efficiency. A translog cost function model is developed, integrating five critical types of energy inputs. The empirical analysis is conducted using a comprehensive panel dataset comprising 26 major sub-sectors within China's manufacturing industry. The results indicate that diesel exhibits the highest own-price elasticity, whereas electricity the lowest. Further analysis highlights the factor substitution relationships and the bias of technological progress through productive service trade and commodity trade channels, providing insights into shifts in energy consumption patterns. Changes in energy efficiency are decomposed into factor substitution effects and technological progress effects via trade channels. The findings reveal the presence of Morishima substitution among three factors. Specifically, productive service trade and commodity imports show a bias towards the combination of energy with labor and energy with capital, while commodity exports are characterized by labor- and capital-biased technological progress. The contributions of factor substitution and the three trade channels demonstrate divergent impacts on energy efficiency improvements across the overall manufacturing sector, as well as within high-energy-consuming and high-tech sub-sectors. Overall, our study enhances the understanding of energy efficiency trends and technological progress in trade-related manufacturing activities, offering a robust foundation for future forecasting.