C21 - Single Equation Models; Single Variables: Cross-Sectional Models; Spatial Models; Treatment Effect Models; Quantile RegressionsReturn

The relationship between government size and economic growth is a major task in the economic literature and this paper is devoted to public expenditure. We empirically study the relationship between public expenditure and economic growth in the European Union. Our approach consists on a quantile regression for the period 2004-2019. The results show a negative and significant impact of total public expenditure on economic growth, with a higher effect in the high tail of the growth distribution. In a more detailed analysis, the study of three large public expenditures, considering the percentage of total public expense, reveals an insignificant effect of health, and a negative impact of education and social protection, higher in the low tail of the growth distribution. Our findings allow better understanding of the effects of public policies on development beyond linear approaches.

This paper explores the role of spatial effects in the innovation processes across 245 NUTS-2 European Union (EU) regions for the period 2008-2012. The goal of the paper is to verify two hypotheses. The first one deals with spatial autocorrelation, i.e., our assumption is that the regional innovation process is not a spatially isolated process but is determined by innovation activities in neighbouring regions as well. Secondly, we assume non-homogeneous responses of innovation output to changes in innovation inputs across groups of regions, i.e., a spatial heterogeneity hypothesis. Patent applications were chosen as a proxy for innovative activity and we considered research and development expenditures and human resources in science and technology as innovation inputs. In order to model the behaviour of innovative activity at the EU regional level, we constructed and estimated a spatial regional knowledge production function model and spatial regime models. The results of the analysis confirm the hypothesis that the regional innovation process is not a spatially isolated process but is also influenced by innovation activities in neighbouring regions, and we have also identified two spatial innovation regimes.

Using the database of logistics performance of countries in the period 2007-2016, we study the convergence of countries in terms of logistics performance across the world, the European Union and V4. In comparison with other beta-convergence studies using the OLS estimation method, we also implement a conditional quantile regression (Koenker and Bassett, 1978). The use of the quantile regression allows us to study the convergence of countries not only from the point of view of the average, but also from the perspective of different quantiles of conditional distribution. We identify countries with different leveles of convergence and quantify different levels of convergence across countries of the world. The results provide empirical evidence of the logistics convergence of countries across the world, and also demonstrate that the countries of the European Union and V4 achieve a higher level of convergence than the average convergence of countries across the world during the review period.

This study investigates the relationship between wages and risks of work-related fatalities in the Czech labour market. To prove this relationship, we followed the theory of compensating wage differentials and the theory of hedonic prices. Using data from 2007 Czech labour survey, we econometrically derived three hedonic wage functions. To estimate mortality risks premiums, two of hedonic wage functions used objective measures of fatality risk, one subjectively perceived risk. We found a positive and statistically significant correlation between wages and occupational fatality risks in all three models. The estimated effect was lower for hedonic function with worker's perception of mortality risks. This result implies the workers themselves underestimate the fatality risks that they are exposed. Based on estimated hedonic wage functions, we computed the value of a statistical life (VSL) which indicates the marginal rate of substitution between compensation and mortality risk. The implied value of VSL is 60 million CZK for the model with subjectively perceived risks and 266 million CZK for the models with objective risks in 2009 Czech crowns.