Tree-based approaches for understanding growth patterns in the European regions

Abstract

Abstract. We run an empirical analysis to understand the main drivers of economic growth in the European Union (EU) regions in the past decade. The analysis maintains the traditional factors of growth used in the literature on regional growth -stage of development, population agglomeration, transport infrastructure, human capital, labor market and research and innovation -and incorporates the institutional quality and two variables which reflect the macroeconomic conditions in which the regions operate. Given the scarcity of reliable and comparable regional data at the EU level, the starting point of the analysis was devoted to build reliable and consistent panel data on potential factors of growth. Two non-parametric, decision-tree techniques, randomized Classification and Regression Tree and Multivariate Adaptive Regression Splines, are employed for their ability to address data complexities such as non-linearity and interaction effects, which are generally a challenge for more traditional statistical procedures such as linear regression. Results show that the dependence of growth rates on the factors included is clearly non-linear with important factor interactions. This means that growth is determined by the simultaneous presence of multiple stimulus factors rather than the presence of a single area of excellence. Results also confirm the critical importance of the macroeconomic framework together with human capital as major drivers of economic growth. This is overall in line with most of the economic literature, which has persistently underlined the major role of these factors on economic growth but with the novelty that the macroeconomic conditions are here incorporated. Human capital also has an important role, with low-skilled labor supply having a higher detrimental effect than the conducive one of high-skilled labor supply. Other important factors are the quality of governance for most developed economies and, in line with the neoclassical growth theory, the stage of development in particular for less developed economies. The evidence given by the model about the impact of other factors on economic growth such as those on the quality of infrastructure or the level of innovation is more limited and inconclusive. The analysis conclusions support the reinforcement of the EU economic governance and the conditionality mechanisms set in the new architecture of the EU regional funds * The authors are grateful to the members of the Economic Analysis Unit of the Directorate General for Regional and Urban Policy, European Commission, for scientific guidance and helpful discussion throughout the analysis. A special thank goes to Beatriz Torighelli, for her essential input in data collection and preliminary data manipulation, and Marina Mastrostefano, who provided insight and expertise that greatly assisted the research and the interpretation of results. The information and views set out in the publication are those of the authors and do not reflect the official opinion of the European Commission.23 24 P. Annoni, ...

Location matters: A spatial econometric analysis of regional resilience in the European Union

“…Of the three, the models specify the impact of the innovation component on economic growth to be the weakest. As striking as it may seem, this finding is again in line with recent analyses on regional growth (Annoni & Catalina Rubianes, ; OECD, ). In particular, the innovation paradox introduced by the OECD () explains the possible reasons for the weak support the model provides for the link between innovative activities and regional growth.…”

“…Neither the infrastructure nor the labour market efficiency component has been found to be statistically relevant in almost any of the model specifications. The reason could certainly be ascribed to the analysis’ limited time spam yet, interestingly, both results are in line with recent economic analyses on regional growth in the EU (Annoni & Catalina Rubianes, ; Crescenzi & Rodriguez‐Pose, ; Rodriguez‐Pose and Garcilao, ).…”

“…Across both spatial regimes, the results infer that a high share of lowly educated people in the work force (at most lower secondary education) is detrimental to growth. As recent contributions (Annoni & Catalina Rubianes, ; OECD, ) find, higher shares of lowly educated people are in fact even more of an impediment to growth than lower shares of highly educated ones, which only underscores the significance of ensuring sufficient levels of education.…”

“…As such, this analysis supplements the basic Solow‐type growth framework with a number of regional explanatory factors that seek to explain the observed differences in growth rates of EU regions between 2009 and 2015. Those factors range from human to physical capital, from population growth to the quality of institutions, from technology uptake to the specialisation of the regional economy (Annoni & Catalina Rubianes, ; Crescenzi & Rodriguez‐Pose, ; Kwok & Tadesse, ; Mankiw et al, ; Mohl & Hagen ; Pescatori, Sandri, & Simon, ; Rodriguez‐Pose, ; Rodriguez‐Pose & Garcilazo ; Rodrik, Subramanian, & Trebbi, ). Conventional growth regressions assume that variables observed at the regional level are independent, but there is an established consensus that regional economic growth rates exhibit spatial dependence (Abreu et al, ; de Dominicis, ; Ertur et al, , among others).…”

Measuring quality of government in EU regions across space and time

“…Correlations sometimes implies causation, and the quality of institutions—and, in particular, the concept of quality of government (QoG)—has recently emerged as “a key factor” for understanding gaps in development across nations (Holmberg et al, ) or European regions (Annoni & Catalina Rubianes, ; Charron & Lapuente, ; Tabellini, ). QoG does not necessarily reflect the formal institutions of a polity, but informal ones, e.g.…”