Vojtech Nosek (p), Pavlina Netrdova

Socio-economic differentiation, its character and evolution, is one of the key topics in human geography. There are many methods and approaches how to assess and measure socio-economic differentiation as well as many characteristics representing this differentiation. Unemployment is a typical socio-economic characteristic, which is being studied in this context. However, due to data limitations, the socio-economic differentiation is often studied on higher geographical levels and patterns of socio-economic differentiation on micro levels are often hidden in regional means. In our paper, we analyse socio-economic differentiation of unemployment on a very detailed, municipal, level in Czechia. Moreover, we use a long time monthly data series (2002-2015). This coherent time series helps us to capture both long term and short term trends and mitigates a typical problem of unemployment data, which often vary according to economic cycles. We identify general tendencies and regularities including seasonal fluctuations and we quantify the importance of different geographical levels. Moreover, we make a typology of municipalities and micro-regions according to patterns of socio-economic differentiation and identify areas with specific regional and local trends and effects. We employ various (spatial) statistical methods including global and local spatial autocorrelation indices and other spatial concentration measures, Gini coefficient of concentration, or Theil index decomposition with geographical standardization separating stochastic and contextual components of differentiation. The combination of primarily non-spatial with spatial methods enables us to come up with innovative results. Besides empirical results, we present a specific methodological framework for studying socio-economic differentiation on a micro-scale.

Esteban Fernández Vázquez, Ana Viñuela (p), Fernando Rubiera Morollón, Alberto Díaz Dapena

territorial inequalities in the European Union are available only at NUTS II or, in the best cases, NUTS III level. However, there are significant internal regional disparities in many social or economic variables that are not observable at this level of spatial disaggregation. It is especially interesting to compare the different dynamics of rural areas located close or far away from the main European metropolitan areas as well as to analyse core versus peripheral patterns among but also within regions.
In this work an alternative econometric procedure to estimate socio-economic local data using official databases is proposed. Household surveys include reasonable measures of income or consumption that can be used to calculate distributional measures, but at low levels of aggregation these samples are rarely representative or of sufficient size to yield statistically reliable estimates. At the same time, census (or other large sample) data of sufficient size to allow disaggregation either have no information about income or consumption, or measure these variables poorly. First, a statistical procedure, based in Elbers et al. (2003) and Tazzoni and Deaton (2009), would be used to combine these types of data to take advantage of the detail in household sample surveys and the comprehensive spatial coverage of a census. The problem of this type of procedures is that when the estimated information is aggregated at regional level the results ate not necessarily consistent with the aggregated data of the household survey. We propose a Generalized Maximum Entropy (GME) approach to the problem of small area estimation that exploits auxiliary information relating to other known variables on the population and adjust for consistency.
As example we use the Spanish case using the micro-data of the Spanish Household Budget Survey and the Spanish Census of Population and Housing, both official statistics elaborated by the National Institute of Statistics (INE). From these two databases we can estimate information at a detailed spatial scale. The potentiality of this approach to the policy-maker is shown by presenting new intraregional analysis that was not possible basing only on aggregated information.

References:
Elbers, C. Lanjouw, J.O. and Lanjouw, P. (2003) “Micro-level estimation of poverty and inequality”, Econometrica, 71 (1), 355-364.
Tazzoni, A. and Deaton, A. (2009) “Using census and survey data to estimate poverty and inequality for small áreas”, The Review of Economics and Stadistics, 91 (4), 773-792.

Milene Simone Tessarin (p), Paulo César Morceiro, André Luis Squarize Chagas

Several factors make firms and, consequently, the productive sectors have a well-defined proximity relationship. With the information technology and telecommunications revolution, technology and knowledge have become even more relevant and explain sector spillovers and their intensity.

The objective of this article will be to evaluate the existence of sectoral neighborhood in the Brazilian manufacturing industry in the recent period. The main question to be investigated is whether the manufacturing sectors have neighboring manufacturing sectors as well as the geographical territories? If so, how does this neighborhood relationship occur? Are there groups of sectors with more intense neighborliness than others?

In this study we measure sectorial neighborhood how the fact of a firm uses the same productive, knowledge and technological base to produce products typical of its sector of origin and other manufacturing sectors. The sectorial proximity occurs due to the spillovers of knowledge and technology that trigger opportunities for firms to establish, on the same productive basis, the production of other sectors.

This study uses an unprecedented and well-disaggregated sectoral neighborhood matrix for Brazil (103 sectors), which was provided by the IBGE through a special request. From this matrix it was possible to explore neighborhood relations through descriptive analyzes and spatial econometric methods. We also created sector indicators of employment weighted by the skills of each occupation related to the innovative and technological activities to statistically capture the sectoral neighborhood pattern. Some sector variables were used as controls, for example, the degree of concentration, indicator of commercial opening, productivity, number of engineers, among others.

We used the Moran's I test, and we used the LM and LM robust tests to choose the best econometric model that confirms the spatial dependence of the sector. We thus find that manufacturing subsectors have well-defined industrial neighbors, that is, they are spatially dependent. All methods attested spatial dependence and were statistically significant. In addition, we built graphs (using Gephi software) highlighting the most intense sectoral neighborhood.

There is a sectoral pattern of between the subsectors of the low- and medium-low-tech industry and also between the high- and medium-high-tech industry sectors. The sectorial neighborhood between different technological levels were less intense. Therefore, the technological origin of a sector is a good sign of its relations of neighborhood with the other sectors. These results may be useful for policymakers by indicating which industrial sectors generate the most spillover effects on the country's productive and innovative structure.

Peter Batey (p)

Regional science is widely acknowledged to have its origins in the 1950s. Isard (2003) traces the developments that led economists, geographers and planners to come together to form the inter-disciplinary Regional Science Association (now RSAI) in 1954 and describes the growth of a distinct body of spatial analytical methods and location theory capable of being used to support urban and regional policy (Isard 1960).
But what preceded this growth in regional science? Is there any indication that formal methods were used by early planners? What role, if any, did quantitative data play in spatial planning? How far were planning policies and proposals supported by an evidence base? Are there signs of an inter-disciplinary approach drawing on a number of social sciences?
At the Vienna Congress last year, I began to piece together some elements of the ‘pre-history’ of regional science methods. I pointed to a number of planning studies from the 1940s in which a rigorous and systematic approach was adopted. These included some important work by economist Philip Sargant Florence in the area of industrial analysis. Sargant Florence, an American by birth who spent most of his working life in England, made significant contributions to the measurement of industrial concentration and the identification of footloose industries that might offer opportunities for reviving the rural economy. As Professor of Commerce at Birmingham University, Sargant Florence was a key figure in the West Midland Group on Post-war Reconstruction and Planning and participated in that Group’s studies of Herefordshire and the Birmingham Conurbation in the 1940s. Sargant Florence’s interests were many and varied and included not only economics but geography, sociology and planning.
In this paper I make an assessment of Sargant Florence’s contributions to the methodology of regional analysis and planning on both sides of the Atlantic.
References:
Isard, W. (1960), Methods of Regional Analysis, Cambridge: MIT Press
Isard, W (2003), History of Regional Science and the Regional Science Association International the Beginnings and Early History, Berlin: Springer.