The effect of research intensity on total factor productivity in OECD countries during 1890–2018: evidence from a new Poisson pseudo maximum likelihood estimation approach

In the literature of total factor productivity (TFP), the roles of several variables in explaining TFP remain largely unexplored, including the impact of research intensity. This paper examines the effect of research intensity on TFP in 15 OECD countries, while providing for financial development and education level during the period 1890–2018. We have used a new Poisson maximum likelihood estimation method, which requires minimal assumptions about the data distribution, to estimate the relationship in the variables. The results reveal that research intensity, financial development, education level and oil crisis have positive impact on TFP, while the two World Wars have negative impact on TFP. In the model featuring all the control variables and dummies, it is observed that for every one percentage point increase in research intensity, TFP grows by 0.086 point. The results further show that the positive impact of research intensity on TFP is more conspicuous in G7 countries relative to non-G7 nations. The sample has been divided into G7 and non-G7 countries and the equations are re-estimated. The empirical results are not materially different from results of the 15 OECD countries. For the sake of robustness, we have also used a new sequential linear panel data estimation approach to estimate the impact of research intensity on TFP, but similar results are obtained. The policy implications of the empirical results are detailed in the paper.