Non‐linearities in regional growth: A non‐parametric approach

Abstract

This paper analyses the determinants of regional economic growth in the European Union adopting a non-parametric approach. Although the local-linear kernel estimator applied does not explicitly take into account the spatial dimension of the data, it is found to be consistent in our context. In addition, the geographically weighted regression turns out to be less efficient. We obtain evidence of a non-linear relationship between regional growth and its determinants in the form of parameter heterogeneity and threshold effects. These non-linearities mainly affect the initial productivity of labour, the human capital endowment and, as a novelty, the level of infrastructures.JEL classification: C14, C20, O18, R11

Heterogeneity and Spatial Dependence of Regional Growth in the EU: A Recursive Partitioning Approach

“…Clearly, integration in the European traffic networks is beneficial for trade and thus for economic development and growth. This variable has been found important, for example, in Sanso‐Navarro and Vera‐Cabello (). A dummy variable for capital regions (capital) is used. This variable has been found significant in Schneider and Wagner () and has posterior inclusion probability equal to one in Crespo Cuaresma and Feldkircher (), in line with the large literature on core‐periphery effects in new economic geography models (compare Fujita et al ., ).…”

Regional Growth, Innovation, and Latent Nonlinear Effects

“…Therefore, kernel estimations will be able to control for spatial dependence when the regressors are close both in the variable and geographical spaces. Furthermore, Sanso‐Navarro and Vera‐Cabello () have shown that the local‐linear kernel estimator is more efficient than the alternative geographically weighted regression method (Brunsdon, Fotheringham, and Charlton, ). These arguments can be added to those in McMillen (, ) to advocate the use of nonparametric methods when dealing with spatial data and motivate their use to analyze the link between knowledge, innovation, and growth in European regions.…”

“…Similarly, Basile and Gress (), Basile () and Basile, Capello, and Caragliu () find nonlinear effects of the initial GDP per capita level and physical and human capital endowments. Sanso‐Navarro and Vera‐Cabello () conclude that nonlinearities mainly affect the initial productivity of labor, human capital, and the level of infrastructures. Adopting a distributional approach, Basile () only assigns a marginal role to nonlinearities in the accumulation of physical capital while Fiaschi and Lavezzi () find that the sectoral composition induces nonlinear growth patterns.…”

“…That is to say, the sample is split in such a way that only the observations of a country are used to obtain the estimates of its regions. Given these findings, and as an alternative way to control for country effects in our cross‐sectional framework, we have carried out the analysis considering the continuous variables as deviations from their corresponding sample country mean (Sterlacchini, ; Sanso‐Navarro and Vera‐Cabello, ). Nonetheless, it is worth noting that differencing in nonparametric and parametric settings is not necessarily undertaken in the same way (Su and Lu, ).…”

“…To the best of our knowledge, these results have only been exploited by Sanso‐Navarro and Vera‐Cabello ().…”