REGIONAL DISPARITIES AND THE REGIONAL DEVELOPMENT: CASE OF SERBIA
Emilija MANIC, Belgrade
Svetlana POPOVIC, Belgrade*
Dejan MOLNAR, Belgrade ***
with 6 fig. and 2 tab. in the text
CONTENTS
Zusammenfassung......
Summary......
1 Introduction......
2 Theoretical frameworks......
3 Regional disparities in Serbia......
4 Discussion and conclusion......
5 Bibliography......
Zusammenfassung
Summary
The paper deals with regional disparities in Serbia, considering demographic and economic regional differences. At the same time, the analyses of those disparities are put in the EU context and context of neighboring country. The authors used statistical instruments to show the extent of Serbian regional disparities, as well as the ArcGIS software to visualize these differences. The results are used to develop rough sketch of possible solution of regional inequalities in Serbia, considering exiting Serbian and EU development documents
1 Introduction
Regional inequality is a salient feature of many countries, both developed and developing. For example, GDP per capita in London is much larger than in the United States, while in Wales it is lower than in Greece. Mississippi (USA) has a GDP per person closer to Slovenia than to many US states, while people in the District of Columbia or Delaware have a higher GDP per capita than most OECD countries. There are many different studies about regional inequalities analysis and considerations of its roots and consequences (YEMTSOV 2003, BADDELEY 2006, KIM 2008, DURO 2004, EZCURRA 2007, HOFER & WORGOTTER 1997, BARRIOS & STROBL 2006, MAGRINI 1999, PETRAKOS 2001, PETRAKOS & SARATSIS 2000, RODRIGUES-OREGGIA 2005, TERRASI 1999, WANG & GE 2004).
The aim of the paper is to show regional disparities in Serbia, both demographic and economic, and to compare it to the neighbouring and EU context. All the research was conducted within the theoretical framework of existing paradigm shift in regional development policy using knowledge of geography and economy. It is very rare to see project in which both, geographers and economists, collaborate in order to sketch some of possible solutions for problems in the area of unbalanced regional development.
2 Theoretical framework
Theoretical explanations for regional disparities are typically based on several different theories: the growth theory (BARRO & SALA-I-MARTIN 1995), the new economic geography (KRUGMAN 1998, FUJITA, KRUGMAN & VENABLES 1999, KRUGMAN & LIVAS 1996, VENABLE 2011), but also some new theories rooted in the area of evolutionary economic geography (McCANN & VAN OORT 2009, BOSCHIMA & FRENKEN 2011). According to the growth theory, the regional inequality tends to rise as growth occurs in discrete locales, but later on inequalities will decline as equilibrating forces such as better infrastructure, technological diffusion, decreasing returns to capital in richer and high-wage areas, diseconomies of agglomeration, and become stronger (WILLIAMSON 1965). A different view has been proposed more recently within the context of the new economic geography school and endogenous growth (ROMER 1986, LUCAS 1988, FISHER 2006, FINGLETON 2011), which argue that increasing returns to scale, advantages of agglomeration of capital and knowledge will tend to perpetuate, or even increase, spatial inequalities.
If we take into account the European experience of solute existing problems related to balancing regional development, regionalization is seen as preferred form of vertical organization of government (state organization) from the three predominant reasons: (i) functional, (ii) economic and (iii) recognition. However, one should bear in mind that the process is an opportunity, but not absolute guarantee.
In Europe, regional development became a relevant topic for policy makers in the 1950s and 1960s. The main objective of regional development policy was to achieve more equity and sustainable growth. The main instruments were redistribution of wealth in the form of centrally controlled financial transfers and wide-scale public investments. Although, this was a period of relatively strong economic growth, fiscal expansion and high employment, several regions were confronted with structural changes in dominant sectors (agriculture, specific industries etc.). During the 1970s and early 1980s regional development policies were adapted to new challenges, focusing on reducing disparities in income and infrastructure and on activities of a social nature. Early adopters were especially those countries that were faced with substantial regional disparities (e.g. United Kingdom, France, Italy, Sweden) and later a major push came from the European level (EU Cohesion Policy, starting from 1988.).
Many international and European studies have shown that regional development policy models applied during the 80ies have produced rather disappointing results, mainly due to slow convergence of lagging regions (BARRO & SALA-I-MARTIN 1991, BOLDRIN & CANOVA 2001). Centrally managed redistribution of subsidies to less developed areas resulted in excessive dispersal of funds, thereby weakening the effect of the scale of public intervention, which ultimately could not produce measurable effects for development and resulted with even bigger legging of undeveloped regions across the EU (PIKE, RODRIGUES-I-IPOSE & TOMANEY 2006).
In the 1990s the disappointing results on convergence became a motivation for changing the paradigm of regional policy in relation to objectives, priorities, tools, actors, and territorial areas of intervention (CAPELLO 2009). Regional development policy has evolved from a model of short-term grants distributed by order from a central government authority to a model of long-term, decentralized development policies aimed at promoting growth in all regions (irrespective of the degree of prosperity) - from dispersed intervention to more selective investments. Development strategies have begun to focus on the endogenous territorial characteristics (instead of exogenous investments and transfers). Various publications therefore suggest that development policies should support growth in all regions, and regions should invest in their own development by mobilizing local resources and funds in order to exploit their specific comparative advantages without excessive reliance on national transfers and grants (OECD, 2009, BARCA 2009, STIMSON et all. 2011).
Over the past decades, many EU-Member States have changed the balance between sectoral and more integrated policies. At the same time, changes have taken place regarding the involvement of regional and local governments in policy design and implementation. This new concept connects three main elements: (i) the place-specificity of natural and institutional resources and of individual preferences and knowledge, (ii) the role played by the (material and immaterial) linkages between places, and (iii) the resulting need for interventions to be tailored to places. The concept is built upon the new paradigm of development policy which main features are based on tailoring interventions to specific territorial contexts and their spatial linkages, and mobilizing and aggregating the knowledge and preferences of local actors (BARCA 2009). Support for regional development policy is therefore to develop internal growth factors, and not merely redistribution of revenue to less developing areas of the country (CAMAGNI 2009). It was clear that development of place cannot be a solely top-down or bottom-up approach, but a multi-governance approach for regional development policy in a European perspective. Whether a sectoral or a territorially, more integrated approach for developing regions would have preference depends largely on the type of policy, the regional conditions and the existing institutional capacity.
3 Regional disparities in Serbia
3.1 Area of study and methodological issues
Republic of Serbia, according to its constitution, consists of two autonomous provinces (Vojvodina and Kosovo and Metohija[1]) and territory outside these provinces called Narrow Serbia. This kind of territorial organization is asymmetric one and over time it has been showed as one of the serious obstacles in transferring jurisdiction from central to province or district level.
Under the Law of territorial organization (SLUŽBENI GLASNIK RS 129-07 2007), the Serbian territory has been divided into 150 municipalities and 23 towns. These 174 units of local government have legal authority and are able to perform certain administration which central government transferred into them. However, under the Law of state administration (SLUŽBENI GLASNIK RS 79-05 2005) it has been constitute an administrative district, too. This unit consists of several municipalities, it doesn’t have any kind of legal authority and it exist only as political unit (29). The last territorial organization was brought two years ago within the Law of regional development (SLUŽBENI GLASNIK RS 51/09 2010), which introduced so called “statistical regions” - NUTS regionalization of Serbian territory (Serbia-north and Serbia –South are two regions within NUTS-1 level, and five more regions are at NUTS-2 level: Vojvodina, Belgrade region, Sumadija and West Serbia, South and East Serbia and Kosovo and Metohija).
Analyzing spatial socio-economic disparities in Serbia, we have faced with several methodological issues. Firstly, the regional statistics in Serbia is really poor (GDP is available only at the regional level for the last two years; there are no continuous time series because of methodological changes, and the established economic development indicator for municipality level). Secondly, the allocation of GDP is another challenge (large number of firms is registered in Belgrade, but their production is located in other regions in Serbia), but also a different price level of the same goods (prices are higher in Belgrade, Novi Sad and other cities, which make their GDP larger than in the case of using the PPP methodology). Thirdly, data for Kosovo and Metohija province is not available (Serbian official statistical institutions do not have any kind of jurisdiction over this territory since 1999, so this area has been excluded from the analysis).
Available data was taken from from national statistical office (census 2002 and the first results of the census 2011, as well as periodical statistics) and were analyze through ArcGIS softwere and some statistical instruments (max-min ratio as a measure of dispersion) for different spatial level (NUTS-2 an NUTS-3 level).
3.1 Demographic disparities in Serbia
According to the 2011 census (first realised results) Serbia has 7.120.666 inhabitants which are for 377.000 inhabitants less than in census 2002 (STATISTICAL OFFICE OF THE REPUBLIC OF SERBIA 2011). The main reason for this is negative natural increase within the whole territory of country (from 0.2 in 1991 up to -4.6 in 2008). The biggest population decrease has eastern and southern parts of Serbia, the parts which had already been faced with this problem for a long period of time. Only three areas in Serbia showed population increase: Belgrade and Novi Sad (positive net migration rate) and municipality of Novi Pazar (positive natural increase for long period of time within Muslim nationality). Serbian population characterized by depopulation process, accelerated process of demographic ageing and migrations within the context of huge demographic regional inequalities.
If we analyze demographic inequalities within the municipality level, it is evident the there very huge considerable spatial inequalities (Fig. 1).
Fig 1: Natural increase rate in the Republic of Serbia, 2010*
Source: Opstine i regioni u Republici Srbiji 2011, Republicki zavod za statistiku (RZS), Beograd, 2012; own analysis
* no data for Kosovo and Metohija are available
Vojvodina shows significant spatial variations of natural increase rate in the municipality level (Novi Sad has positive natural increase, but peripheral municipalities on the east and north show very negative trends, between -10 and -20). However, the worst situation is in East and South Serbia region, where only two municipalities have positive natural increase (Bujanovac 1,9 and Presevo 3,1) and the greatest number o them have negative trends (most of them have over -10 with extremes such as Crna Trava -47,1 or Gadzin Han -20,5). Depopulation process brought Serbia in rang of the most aged population in Europe and world.
However, the existing regional inequalities in ageing rate and existing demographic trends are the result of the continuous process of emigration from certain parts of Serbia towards (Fig.2).
Fig. 2: Neto migration saldo in the Republic of Serbia, 2010
Source: Opstine i regioni u Republici Srbiji 2011, Beograd, 2012; own analysis
* no data for Kosovo and Metohija are available
If we put aside the migrations caused by war in the ex-Yugoslav areas (refugees from war affected areas, internaly displaced people - IDP), the biggest migration flows in Serbia were from rural towards urban areas, from South towards North of Serbia and from all over Serbia to Belgrade region! According to the results of previous population census, in 2002, the biggest negative net migration rate had border municipalities in Serbia (eastern part of Vojvodina region, eastern and southern part of East and South Serbia region, western part of Sumadija and West Serbia region, between -10 and -20). It is evident that young people and women are leaving these municipalities so the population that is left behind is not able to produce simple reproduction of the population.
Beside the unsatisfied old and sex population structure in Serbia and obvious regional disparities, the education structure of the population is very bad, too. Serbia in general has very small share of high educated people (around 5%), but the lowest share show great concordance with those municipalities with negative population increase (negative natural increase and net migration rate). Regional analysis show great spatial misbalance in this indicator (the highest shares of high education people have university centers: Belgrade, Novi Sad, Subotica, Kragujevac and Nis), which enable these centers to develop even more end to increase existing disparities. In such way the undeveloped municipalities and regions are lagging behind and not having any chance to produce enough human capital which could help them to pull them out in the future.
3.2 Economic disparities in Serbia
Serbia is characterized by large differences in the level of economic development between regions (NUTS-2 level) and municipalities (NUTS-3 level). Spatial analysis confirms high regional disparities at all levels in high unemployment, decreasing economic activity, decreasing number of inhabitants in undeveloped areas.
Regional and intraregional inequalities in the Republic of Serbia officially are presented like undeveloped area, developed centre and not sufficiently developed periphery (REPUBLICKI ZAVOD ZA RAZVOJ 2009 and 2010), and established within the unified development list of regions and local territorial units (municipalities) (SLUŽBENI GLASNIK RS 69-11 2011). This list classifies all municipalities as: areas with the level of development above the Republic average (19 municipalities), areas with the level of development of the 80-100% of the Republic average (33 municipalities), not sufficiently developed areas, with the level of development of the 60-80% of the Republic average (47 municipalities), undeveloped areas with the level of development below 60% of the Republic average (46 municipalities) and devastated areas, with the level of development below 50% of the Republic average (27 municipalities), (Fig. 3)
Fig. 3: Serbian municipality level of development according to the average of the Republic of Serbia, 2010
Source: Opstine i regioni u Republici Srbiji 2011, Beograd, 2012; own analysis
Considering Gross Domestic Product (GDP) as one of the key economic indicators, NUTS-2 level regions in Serbia has been analyzed.
Belgrade region has the largest proportion of Serbia’s GDP (40%) and it’s positioned as the only developed region in Serbia, while all other regions are below national average (Vojvodina 26%, Sumadija and West Serbia 19,5% and South and East Serbia region with the smallest proportion of GDP, 14,50%). The same tendency shows the GDP per capita (GDP p.c.): Belgrade with 180% of national average, Vojvodina (95%), while other two regions are significantly underdeveloped (South and West Serbia 63%, Sumadija and Weste Serbia 71%).
Fig. 4: Regional and intra-regional economic disparities, 2010
Source: Statistical Yearbook - Municipalities and Regions in Serbia, Statistical Office of the Republic of Serbia, Belgrade, 2011, own analysis
Belgrade region has the highest share in total Serbian output considering Gross Value Added (GVA)[2] (more than 4 times higher than the production in the South and East Serbia region). As shown by the max/min ratio, regions are not homogenous. In Juznobacki district (Vojvodina), which is the most developed, the production is more than 6 times higher than in the less developed Severnobanatski district. Similar differences exist in South and East Serbia region, while in Sumadija and West Serbia region shows lower differences.
Considering the employment as one of the development indicators, it is noticed that one third of the total employed persons in Serbia works in Belgrade region and only 18% are employed in South and East Serbia region. Knowing the huge demographic disparities, the number of employees per 1000 inhabitants has been chosen as more reliable indicator of regional inequalities. However, the tendency remains the same: the highest “density” of jobs is in Belgrade, and the lowest is in the South and East Serbia. Max/min ratio of the number of employed persons at the district level shows significant differences, but the differences at the municipalities’ level are even greater (Fig.4). It is interested that one of the highest disparities exist in Belgrade (max/min ratio between Belgrade municipalities is 13,6; in the municipality Grocka employed are 143 persons per 1000 inhabitants, while in Savski Venac 1940).
The given regional disparities with the same trend among regions could be seen, also, using the average salary as development indicator (in Belgrade it is 124% of average salary in Serbia, in Vojvodina close to the Republic average, while in other two regions the salary is close to 85% of the Republic average). The largest intraregional differences in this indicator are in the region of South and East Serbia (max/min ratio at the counties level is 1,4), followed by districts in Vojvodina (max/min is 1,2). At the municipal level exist even larger differences (in the region South and East Serbia employee in Pozarevac earns on average 2,3 times higher salary than employee in Vladicin Han; in Belgrade and Sumadija and West Serbia max/min ratio is 2,1 and the smallest differences between average salary are among Vojvodina municipalities, with the ratio 1,8).