DETERMINANTS OF TOTAL FACTOR PRODUCTIVITY ANALYZED BY DECISION TREE METHOD: THE CASE OF USA 1991-2020

Abstract

Technological developments have significantly impacted value creation in the manufacturing industry. It has been emphasized that the accelerated technological changes since 1990 and the associated income increases are mainly due to the growth of labour, capital and other production inputs, i.e. total factor productivity (TFP). Therefore, recent development and growth accounting studies have revealed that TFP is a vital source of economic development. This study uses the decision tree method from data mining methods to analyze the factors affecting the total factor productivity in producing durable goods in the US economy between 1991-2020. Classification and Regression Tree (CART) algorithm was used for the analysis. There is no decision tree method in TFP analysis, and this study aims to fill the gap in the literature. The findings obtained as a result of the analysis support the literature. According to the decision tree result, the ten best scenarios are given. In addition, the increase in the number of patents, employment of researchers and the share of R&D expenditures in GDP significantly affect the growth in TFP.