Roberto Basile (p), Gloria Cicerone , Lelio Iapadre

In this paper we analyze the role of economic complexity as a driver of regional labor
productivity dynamics in Italy. Economic complexity is expressed in the composition of an
economy’s productive structure and reflects the emerging combination of the multiplicity of
knowledge embedded in it. Measures of economic complexity (ECI) come from the structure
of the network connecting economies to the products they export. We assess the impact
of ECI on the distribution dynamics of labor productivity by combining growth regression
analysis with conditional density estimates. Counterfactual analysis results suggest that ECI
plays a key role in the observed tendency to polarization of regional labor productivity.

Federica Maggi (p), Giulio Bosio , Anna Maria Falzoni

During the last decades, a process of radical transformation has changed labor markets in developed countries, challenging the process of occupational upgrading as well as the nature of jobs (Autor 2015; Goos et al. 2009). More specifically, employment structure polarized, with a wane of routine occupations accompanied by a simultaneous growth of abstract and manual occupations (the non-routine ones). This process, defined as job polarization, implies a substantial change in the demand for task on the labor market. The most accredited explanation related employment polarization with the diffusion of computer-based technology that allows to replace workers in the routine jobs (Goos et al. 2014).
On the other hand, globalization and international trade could reflect alternative forces able to challenge the evolution in labor demand, both at national and local level. In this perspective, the role of economic specialization as an engine of economic performance has been extensively debate in the empirical research, giving particular relevance to the economic complexity index (ECI) (Hausmann and Hidalgo, 2011). This measure, considering the knowledge intensity of exported products, is useful to predict the economic growth at the local level.
However, with few exceptions, job polarization has been investigated at the national level, while the empirical evidence for several European countries seem to suggest a relevant heterogeneity in the degree of employment polarization both across industries and local areas.
In order to fill this gap, this paper investigates to what the economic complexity index matters in capturing the relevant heterogeneity in the degree of job polarization across the local labor market in Italy. Focusing on the period after the Great Recession (2009-2018), we examine if there is any nexus between economic complexity and the job polarization in the Italian provinces. The aim of this research is to understand whether more competitive provinces experienced a stronger polarization in their employment structure.
The lack of previous literature on this topic and the Italian context make for an interesting addition to existing studies on this topic. Matching data at the province level (NUTS 3) for the exports (ISTAT- COEWEB) and the Italian labor force survey (ILFS), we exploit a pseudo-panel approach in order to account for potential unobserved heterogeneity at the provincial level. To further disentangle the influence of surrounding provinces, we also adopt a spatial approach that allows capturing spatial interdependencies across local labor market and reflects a novelty in the literature on job polarization.

Adam Cox, Monica-Mihaela Tudor (p), Menelaos Tasiou, Dan-Marius Voicilas, Mihai-Alexandru Chitea

Whilst recent research has had a keen focus on the concept of a Smart City, less focus has been given to “smartness” in the regional context. Under the umbrella of the European Commission 2020 Strategy, smart specialization is a central pillar in of Cohesion Policy which is based on the principle of Smart Growth (Regional Policy Contributing to Smart Growth in Europe [COM(2010)553]). Accordingly, European regions are called to identify the sectors, technological domains, and major areas of likely competitive advantage to focus their regional policies (Mc Cann & Ortega-Argilés, 2015). Firstly, this paper aims to construct a unique measure “smartness” at regional (NUTS II) level using Romania, as one of the most rural countries of the EU, as a case study. Secondly, this paper then tests the regions’ capacity to deliver on economic growth convergence. Here there is a particular focus on the development regions in Romania, which typically have a weak negotiation position in terms of regional design and implementation of Cohesion Policy. Results show that Romania’s capital city NUTS II region, Bucuresti, is the “smartest”. Western Romania is considered a medium level of smartness and the lowest smartest is seen in the eastern and southern Romanian regions. We find a statistically significant correlation between regional smartness and the ability of regional economies to bridge the gaps compared to the EU average. These results provide evidence to support the notion that enhancing the underlying indicators of ‘smartness’ would support a region’s economic resilience and development.

Acknowledgment: paper is supported by Horizon 2020 research and innovation programme under the project: Perception and Evaluation of Regional and Cohesion Policies by Europeans and Identification with the Values of Europe (PERCEIVE).

Lucian-Liviu Albu (p), Ada-Cristina Marinescu

Following other studies demonstrating empirically the complex road to convergence in EU, based on an updated analyse, we highlight as a quasi-general rule: a country starting at a relative small level of income per capita (far on the left side at that moment from the average of its group level) will strongly converge to the group average but in this way it will be registered a process of divergence among its component regions. A main problem is when the regional divergence will reverse in a regional convergence process. In our study we propose a methodology to explain this phenomenon of switching from regional divergence to regional convergence doubled by a simulation model. As application, we are using few levels of grouping (digitalisation). Thus, firstly, EU (27 countries after Brexit) is divided in three conventional groups of countries. Secondly, each country is divided in its component regions, according to the NUTS 2 Eurostat database. Thirdly, each region in a country is divided in its smaller territorial units (counties), according to the NUTS 3 Eurostat database. Finally, considering the period starting in 2000, we present a resulted typology of countries and respectively of regions and counties in EU by using two criteria (changes in position against the average income per capita in EU and respectively existence of a process of convergence/divergence among components).