Search results

Recent advances in stochastic frontier analysis (SFA) suggest two alternative approaches to account for unobserved heterogeneity and to distinguish between persistent and transient inefficiency. The first approach is the generalized true random effects (GTRE) model, and the second approach is an autoregressive inefficiency (ARI) model. This study compares them to highlight whether they capture similar inefficiency aspects.

Using recent methodological advances in SFA, the authors estimate the GTRE and the ARI models using a Monte Carlo experiment and two real datasets from two industries (banking and agriculture).

The authors find that the two models provide quite different results in terms of inefficiency persistence and overall inefficiency (combination of transient and persistent inefficiency), regardless of the dataset considered.

The study findings suggest that researchers should be careful when referring to these two models because they do not capture the same inefficiency aspects, even though they have the same conceptual basis. This work is a warning about the empirical aspects of the persistent and transient efficiency framework, in order to convey a consistent story to the reader on firms' performance.

Even though they are used in a large number of studies, the present paper contributes to the productivity and efficiency literature by providing the first comparison of the GTRE and the ARI models.

Research and development (R&D) is increasingly considered to be a key driver of economic growth. The relationship between these variables is commonly examined using linear models and thus relies only on single-point estimates. Against this background, this paper provides new evidence on the impact of R&D on economic growth using a machine learning approach that makes it possible to go beyond single-point estimation.

The authors use the kernel regularized least squares (KRLS) approach, a machine learning method designed for tackling econometric models without imposing arbitrary functional forms on the relationship between the outcome variable and the covariates. The KRLS approach learns the functional form from the data and thus yields consistent estimates that are robust to functional form misspecification. It also provides pointwise marginal effects and captures non-linear relationships. The empirical analyses are conducted using a sample of 101 countries over the period 2000–2020.

The estimates indicate that R&D expenditure and high-tech exports positively and significantly influence economic growth in a non-linear manner. The authors also find a positive and statistically significant relationship between economic growth and greenhouse gas emissions. In both cases, the effects are higher for upper-middle-income and high-income countries. These results suggest that a substantial effort is needed to green economic growth. Internet access is found to be an important factor in supporting economic growth, especially in high-income and middle-income countries.

This paper contributes to underlining the importance of investing in R&D to support growth and shows that the disparity between countries is driven by the determinants of economic growth (human capital in R&D, high-tech exports, Internet access, economic freedom, unemployment rate and greenhouse gas emissions). Moreover, since the authors find that R&D expenditure and greenhouse gas emissions are positively associated with economic growth, technological progress with green characteristics may be an important pathway for green economic growth.

This paper uses an innovative machine learning method to provide new evidence that innovation supports economic growth.