Lucian-Liviu Albu (p)

Last decade there is an increasing preoccupation of researchers and policy makers on the regional convergence in EU. The study is starting with a short analyse of convergence/divergence at the level of whole EU and inside of some its conventional groups of countries (three groups: new eastern members, southern members, and old western members). Then, we propose a methodology to classify countries in EU in four distinctive groups resulted by the combination of two criteria: changes in position of a country related to the average income per capita in EU and respectively regional convergence or divergence inside of this country. Thus, our study could be viewed as focussing on the identification of main so-called clubs of regional convergence inside of EU. Moreover, by studying the differences among such identified clubs or groups, a number of parameters characterising potential transition from a behavioural regime to another could be estimated. Finally, we propose for future research a general unifying model to describe transitions along with the economic growth process, and a corresponding calibrated model to be applied for simulation on available regional data in EU (NUTS 2 data). The results and conclusions of the study could be useful for future research on potential convergence trends and for decision makers in order to orientate their regional policies and corresponding measures to a more efficient allocation of available resources.

Calin Arcalean (p), Gerhard Glomm , Ioana Schiopu

In contrast to most of the literature on optimal urbanization that relies on the existence of agglomeration externalities of different types, in this paper we study a completely different mechanism. We show that the lack of complete markets across physical space – in other words the existence of non-tradable sectors – can drive a wedge between centralized and decentralized allocation of workers across regions in the absence of any externalities in production or consumption.
JEL codes: R12, R13

Johan P. Larsson (p), Martin Andersson

The debate over the relationship between productivity and the density of economic activity has largely been based on analyses where productivity differences between cities and sparser areas are attributed to differences in their density. Therefore, much of the empirical evidence of the role of density comes from productivity effects driven by large dense cities. This is particularly the case in recent work on within-city neighborhood density effects. However, there are reasons to believe that agglomeration economies, relatively speaking, are at least as important in sparser areas. For example, the role of dense towns, or neighborhoods, in the countryside could, in functional terms, play the same role as a dense neighborhood do in a large city. In this paper, we analyze the magnitude and spatial attenuation of agglomeration spillovers in different spatial contexts. We exploit geocoded Swedish data aggregated unto square grids of various sizes and study how the elasticity between density and productivity varies across spatial contexts. We also assess heterogeneity across spatial contexts in the rate at which agglomeration effects dissipate with distance. By aggregating matched employer-employee panel data over the 1990-2015 period unto these grids we document quantitatively important elasticities, indeed comparable to those in metropolitan areas, even in sparse rural areas. In fact, our estimates reveal sizeable neighborhood effects even in rural areas, and that those effects exhibit lower depreciation with space compared to urban areas.

Izabella Szakálné Kanó (p)

Nowadays, we cannot speak about regional development without mentioning agglomeration economies. On the one hand, in regions with diverse economic environment, urbanization economies provide firms with skilled workers and they can also gain advantage from knowledge spillovers of other firms. On the other hand, in less diverse regions, localization economies can offer advantages, which the firms benefit from in form of specialized local labour market, specialized local value chains and intra-industry knowledge spillovers. Especially in case of knowledge intensive industries, it is crucial to have highly educated labour in place.
Knowledge-intensive industries have attracted a great attention nowadays in researches because of their contribution to the development of knowledge driven economy. They generate positive effects on the regional economy and have increasingly high importance in less developed regions, like Hungary.
Several different approaches can be found in the literature measuring the specialization of regions and the concentration of industries. In order to quantify them, we apply Local Moran index (LISA) based on Location Quotient. The human capital, as the key input of knowledge intensive industries, can be measured also by a number of technics, among which we use proportion of tertiary educated population.
Our investigation is based on city-regions and subregions level. In order to apply regional development strategies in regions, one has to consider nodal regions, i.e. functional regions established from labour commuting zones with a powerful centre: the 23 city-regions. Labour commuting zones often extend beyond the borders of subregions (175), but latter are still well applicable for investigation of concentration.
In our study we aim to answer following questions:
1. Which of the considered industries were mostly knowledge intensive in sense of best resembling the spatial pattern of tertiary educated population?
2. In case of industries, which had the closest spatial pattern, what is the extent of spatial lag in effect of human capital on their pattern.
3. What is the most likely time lag between the change in human capital and change of spatial concentration of individual industries?
According to our expectations knowledge intensive service sectors and manufacturing sectors, like Manufacture of basic pharmaceutical products and Manufacture of computer, electronic and optical products were the sectors, which were spatially closest to the pattern of tertiary educated population. Based on one of our former study, we expect a 1-2 year lag in case of high-technology industries and a longer time lag in case of other industries.

Pasquale Commendatore (p), Ingrid Kubin , Iryna Sushko

See extended abstract