The R&D Spending Effects on the Manufacturing Firms’ Productivity Evidence from Poland using the PSM and MDM Experiments

Abstract

This research paper seeks to establish substantive evidence on the impact of R&D expenditures on productivity in SMEs in the formal private manufacturing sector in Poland, using the World Bank Enterprise Survey (WBES) 2025 data. Poland had been chosen as a case study principally due to the fact that it represents a major transitional economy in the Central Europe region, and a significant manufacturing hub in the European Union area.

The novelty of this paper is inspired by the argument made by (King and Nielsen, 2019). This is where they suggested that the Propensity Score Matching PSM, namely Nearest Neighbour matching, could approximate a low-standard experimental design, and could ignore much of the potentially useful information without efficient use, leaving us with higher imbalance, model dependence, and ultimately bias. Thus, these developments in the matching methods suggested that the conclusions drawn from PSM analysis are best supported by a second estimator, such as Mahalanobis Distance Matching MDM, which has the property of double robustness, reduces imbalance, model dependence, and bias. Therefore, both the completely randomised experiment procedures by PSM, and the fully blocked randomised experiment by MDM are recommended for more confidence and reliability in the obtained results.  

The findings show that there is statistically significant impact of R&D spending (Treatment) on firms’ peroductivity as the (Outcome) variable using the Kernel-based matching and Mahalanobis distance matching, but the effects appeared to be statistically insignificant using the nearest neighbour matching procedure. These results suggest that Poland should adopt more effective strategies to promote R&D expenditures in the manufacturing sector owing to the positive impact of R&D spending on productivity in the firms operating in this important sector.