In this study, we examined the impact of the Fourth Industrial Revolution’s new technologies on the economic performance of enterprises using a propensity score-matching approach, grounded in a potential outcome framework, to mitigate endogeneity issues in estimating the effects using regression. Propensity score estimation performance was bolstered by employing a machine-learning algorithm, the random forest. Data from the Survey of Business Activities in Korea, collected through comprehensive enumeration, were used to enhance the results’ generalizability. Findings revealed that enterprises adopting new technologies experienced an approximately 40% surge in annual sales compared to those that abstained. Effect magnitude varied based on enterprise size. For small-and mediumsized enterprises, the increase was approximately 43%; for large enterprises, 77%. Thus, large enterprises appear to have an advantage over smaller ones in utilizing new technologies. This study is significant for its application of the latest research methodology in estimating economic effects of Fourth Industrial Revolution techn ologies on enterprises and identifying policy implications.

Using a range of macroeconomic and financial variables as predictors, we examined the out-of-sample predictability of iron ore prices through machine learning (ML) models. Our result shows that traditional factor models or/and ML models such as random forests and extra-trees, outperform the linear regression. For longer forecasting horizons, the Elastic Net model demonstrates superior accuracy. This suggests that choosing informative predictors is crucial for improving the forecastability of iron ore prices.

This study analyzed carbon emissions in the carbon-intensive manufacturing sector of the Chungcheong region and their relationship with productivity. Industry-specific carbon emission coefficients were derived using regional input-output tables, and factors influencing emission changes within and outside the region were examined using the extended Syrquin method. The System GMM estimation method was employed to explore the impact of carbon emissions on labor productivity. The analysis revealed that direct emissions had the largest share, with some industries affected by regions outside Chungcheong. Changes in emission coefficients were a significant factor, and regions outside Chungcheong had a notable influence. Labor productivity in the Chungcheong region’s manufacturing sector remained unaffected by carbon emissions, while regions outside Chungcheong saw improvements with decreasing carbon emissions.

The aim of this study is to analyze effects of portfolio changes in public R&D investments on the Korean economy. To that end, we developed a computable general equilibrium model that enables industry-specific simulations. This study is distinguished from others in that it focused on public R&D investments of two major sectors in Korea: steel and semiconductor industries. The simulation results show that an increase of public R&D investment in the steel industry has an positive effect on the whole economy. On the other hand, an increase of public R&D investment in the semiconductor industry has an positive effect to some extent, but give a negative effect on the national economy. This implies that there is an optimal point of public R&D investment in each industry.