CONVERGENCE IN THE DEVELOPMENT REGIONS OF ROMANIA
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
CONVERGENCE IN THE DEVELOPMENT REGIONS OF ROMANIA Associate Professor Ph.D SÎRGHI Nicoleta West University of Timisoara,Faculty of Economics and Business Administration [email protected] Lecturer Ph.D PĂREAN Mihai West University of Timisoara, Faculty of Economics and Business Administration [email protected] Assistant Lecturer OŢIL Maria West University of Timisoara, Faculty of Economics and Business Administration [email protected]
Abstract
The present paper consists of two parts. Part one focuses on regional growth theories which have registered a gradual evolution, marked, in the beginning, by the analysis of regions as non-spatial elements of national economy. Part two suggests an econometric testing of the two types of real convergence: beta convergence and sigma convergence in the development regions of Romania. Thus, the period of time required by the development regions of Romania to reduce the discrepancies in GDP per inhabitant compared to Bucureşti Ilfov region as well as to the EU-27 average is determined.
Convergence is a dynamic process based on the application of socio-economic policies designed to reduce disparities among regions and countries in a given space. It is achieved by implementing the structural policy in order to obtain sharp economic growth parameters in the peripheral regions which experienced a period of economic downtrend or didn’t manage to attain the economic performance of the region to which they belong. Regional growth is considered of utmost importance when referring to the real economic situation of a development region.
the comparative costs, etc.) which determine how the given economic growth rate will be distributed among the system’s regions. Within these models, a region’s growth will always bring about the other’s downfall. Many of the traditional regional growth theories (cumulative causality, neoclassical theory, etc.) consider regional growth as competitive growth. The national growth rate is determined exogenously, while regional economic analysis is concerned only with the distribution of growth among regions. The need to include in these variable spatial models is reduced since every region is considered an economic sector. However, according to the generative growth models, the regional dimension is far more accentuated, as these consider national economic growth rate as the resultant of regional growth rates. Consequently, the entire growth is spatially oriented, in other words the growth of any national economic
Key words: convergence, development, sigma convergence (σ), beta convergence (β). 1. Introduction The concept of regional growth has evolved gradually in time and was marked in the beginning by the analysis of regions as non-spatial elements of the national economy. One of the consequences is that regional growth rates are considered to derive from the national growth rates, the propulsive impact of regional growth on the national growth rate being ignored. This way of approaching regional growth emphasizes the idea of interregional competition, hence the competitive growth concept. The competitive growth models imply that the attainable rate of national growth is known, thus examining the forces (the advantages and disadvantages of location, the relative potential of the market,
727
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
sector has its origins in a certain location. Regional growth performances can be improved without producing adverse effects on the growth rate of the neighbouring regions. Growth generated by the innovation process can be integrated into the following context: the agglomerations and the spatial proximity of activities in certain towns or regions may lead to an innovation rate above that registered in the absence of agglomerations. Similarly, the changes manifested within the intraregional distribution of productive inputs, enabled by an efficient intraregional transport system for example, may increase both production efficiency and regional growth rates. The importance of spatial impact on regional growth is overlooked in the case of competitive growth models. The phenomenon that allows intraregional spatial efficiency to have a feedback effect on the aggregate growth rate is called generative growth. These brief considerations reinforce the necessity of taking into account the spatial component when elaborating the regional economic growth theory and the urban-regional analysis. In practice, it is reflected in the careful study of the implications generated by the adoption of certain regional development policies which differ depending on the amplitude of political power interventions, the centralization or decentralization of regional policy implementation, the macroeconomic or microeconomic orientation, forecast measures having an inter and intraregional field of action, the focus on the redistribution of workforce or capital, the exogenous sources or the endogenous development potential, the upgrading process, the expansion of the tertiary sector, or the traditional sectorial measures. At the same time, a region can present a reduced overall output growth and a rapid increase of the output per capita if the immigration process in that region registers significant quotas during the study period. The most commonly encountered are: a region’s overall output growth, growth of the output per occupied person and per capita. In its turn, the output can be determined by dint of: total (gross) output of a region, the regional GDP, the regional net domestic product, etc. The problem is finding the most adequate way to measure regional growth. This also depends on the purpose of
the measure. Therefore, overall output growth is used to indicate the growth of the regional production capacity, which partially depends on the extent to which the region attracts capital and labour force from other regions. The growth of the output per occupied person is often used as an indicator for the modifications in regional competitiveness by means of productivity growth. The growth of output per capita indicates changes in the economic prosperity at region level. Another controversial issue in relation to the regional growth disparities refers to the medium and long-term vision concerning the regional growth consequences. Thus, according to the neoclassical model, which highlights the role of offer, regional growth leads to convergence in the regional socioeconomic development, while conforming to the models based on the postKeynesian approach to demand (the model based on the export potential, the cumulative model), regional growth underlines divergence. Their complementary approach was outlined against the two visions on regional growth consequences, depending on the general level of development, the economic policy options of a country and issues specific to different regions of the territorial system. 2. Literature review The liberal doctrine of development is based on three major concepts: the economic growth pole, the industrial complex and the industrial district. The growth pole concept (growth pole strategy) was developed by Francois Perroux. Since growth pole level is accompanied by a superior innovative and competitive capacity, the advanced areas may generate a continuously polarized development. In the absence of barriers concerning the workforce and population migration, the appearance of growth poles may be associated with important population displacements which represent another demographic issue[1]. Generally, the direction of these displacements is from the insufficiently developed regions towards developed ones and concerns the young and middle-aged population as well as university graduates. As a result, the developed regions impose a series of filtering processes on the input flows of workforce. Such phenomena lead to a mainly qualitative depopulation of
728
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
the insufficiently developed countries, influence the rhythm of demographic evolution (marriage and birth rate), emphasizing demographic ageing and producing a decrease in regional competitiveness. The developed regions’ filtering of activities and the competitiveness loss of the less developed regions, together with the qualitative depopulation may lead to economic duality. According to liberals, the main causes of regional development disparities are: the lack of entrepreneurship education and excessive state interventions. The revival of the insufficiently developed regions can be achieved by means of labour market deregulation, free functioning of markets, and privatization of public enterprises. According to the Keynesian interventionist approach, the main causes of regional disparities are market insufficiencies and capital concentration. The solution for reducing disparities would be the intervention on behalf of public, national, regional and local authorities by means of public investments into the socioeconomic infrastructure [2]. The incentive practice may be a way towards regional development. The main incentives are: tax decrease; land lease; subsidized credits; employment growth incentives; facilities for SMEs. A strongly controlled economic policy may prevent the territorial and economic disparities in the short run, leading instead to mainly political artificial privileges, to the disadvantage of public and private sector. This causes hidden economy development, enlarged deficit, public debt growth, disinvestment, increase in inflation and decrease in income growth and tax basis. In conclusion, interventionism would produce stagnation and long-term increase in inequalities, i.e. exactly the opposite of initial intentions, thus imposing financial liberalization measures and restrictive monetary policies. At the beginning of the ‘70s, regional development mainly focused on the eradication of poverty and development of small towns. Thus the idea of development by means of central growth was abandoned. The political problems at that time indicate that the national growth objectives (regional) differed from the rural development ones in terms of anticipatory effects. Therefore, the necessity of studying the implications of
rural development policy based on the growth poles was acknowledged. Subsequently, other policies and concepts were developed referring to the urban centre development in conjunction with rural development. Initially, it was the states that undertook to reduce regional disparities. Within a short period of time, the impact of implementing regional development programs achieved by the state and mainly based on fiscal incentives proved to be quite ineffective [3]. The revival of regionalist movements in the ‘70s, the implementation of administrative reforms in a number of countries that passed laws related to regional level, the European Community enlargement and the increasing transfer of duties towards community institutions contributed to the radical change in the regional development phenomenon. Now, regional development has different meanings in the European countries. Within the framework of the European Union, few states analyse regional development in terms of the major problems posed by economic development of internal territories compared to the economically developed and privileged regions from the geographical, historical and resource endowments point of view, etc. Others consider the regional development approach to ensure the effective and active participation of all citizens to the economic, political and social development of the country. Thirdly, it implies the development of the territories far from the economically developed centre. However, in all the abovementioned situations, regional development is an already traditional and consistently promoted government policy sustained by community funds. In Central and Eastern European countries, regional development is a relatively recent phenomenon, intensely promoted by dint of the democratization process and the intensification of socioeconomic development while maintaining and supporting the territorial integrity and expanding the interstate collaboration. The focus is on the creation of units, the so-called territorial regions, distinct from the existing intern administrative-territorial state structure and on the general balanced socioeconomic growth of these structures. In a short period of time new priorities have been established
729
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
for territorial development, objectives have been modified, an institutional framework for the financing of development programs was created and the instruments for regional development policy implementation diversified so as to concretely underline the development level and to offer an overall report of territorial units for regional statistic accomplishment thanks to the implementation of the Nomenclature of Territorial Units for Statistics (NUTS).
takes place when the economic growth gap is widening. GDP per inhabitant at purchasing power parity represents the classic real convergence indicator [4] Solow’s economic growth model demonstrates the possibility of catching up with richer countries as far as income per inhabitant is concerned. The neoclassical theory based on this model upholds the possibility of achieving the economic convergence process among countries. The fundamental hypothesis of the neoclassical model is "the dependence of convergence (catching-up process) on the return on capital, on its generally decreasing trend. Capital growth will correspond to a less than proportional return on capital increase.” [9]. The poor countries, with a low GDP and physical capital per inhabitant, grow faster compared to richer economies with a higher GDP and capital per inhabitant. The neoclassical economic growth theory indicates that all economies characterized by the same basic parameters will attain an equal level of development, regardless of their initial state. As a result, real convergence hypotheses were formulated based on the convergence theory conclusions: absolute and conditional convergence hypotheses. According to the absolute convergence hypothesis, the farther a country is from its long-term equilibrium, the higher the income per inhabitant; the initially poor counties will evolve faster than the initially richer ones. Absolute economic convergence appears when there is a reverse relation between the initial level of income and the economic growth rate, i.e. ”the economic growth rate decreases as the income level reaches the steady state”[10] . The conditional convergence hypothesis implies that if certain structural parameters differ, i.e. the analysed countries or regions do not possess the same structural characteristics, then „each of them will converge towards its own limits” [1]. Although growth rates may be higher in developed countries than in richer countries, the development differences remain significant (they are maintained and even accentuated, leading to income divergence). This convergence type implies that all economies with different initial capital stock per capita have the same savings rate, similar technologies, as well as the same population growth rates. In case of non-
3. Hypotheses and convergence types Convergence represents a dynamic process based on the application of socioeconomic policies designed to reduce disparities among regions and countries in a given space. It is achieved by means of structural policy implementation in view of obtaining sharp economic growth parameters in the peripheral regions which experienced a period of economic downtrend or didn’t attain the economic performance of the region to which they belong. Another definition of convergence pertains to the increase in similarities and economic performance of regional and national economies in a given space. Convergence is seen as a preliminary condition for integration because while policy implementation and creation structures converge, the integration process together with its strategies and the establishment of common functional institutions are more easily achieved. Therefore, convergence means working towards the same goal, evolving in order to reach a target. The convergence problem lies in “establishing the extent to which economies with different initial output levels register a growth that allows them to finally reach equal living standards” [1]. The convergence process has in view the approximation of the economic, social, monetary, fiscal and performance indicators of regions and countries, the reduction of development disparities, the assurance of financial and monetary stability in all the countries as well as the compatibility of institutional and regional administrative structures and mechanisms. We can speak about real economic convergence if poor countries’ economic growth rate surpasses that of the richer states, reducing income disparities among them. The economic divergence process
730
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
fulfilment of conditions, convergence cannot occur. Later, Mydral (1957) replaces the stable equilibrium hypothesis with the circular cumulative causation which determines the divergence phenomenon [13]. The hypothesis explains the persistence and widening of international and interregional differences at development level. Regional and international inequalities in different countries are perpetuated and even widened by means of capital movement, human capital and workforce migration, and the exchanges of goods and services. It’s the polarisation trend (agglomeration) in the conditions of economic and monetary integration that accentuates the level of disparity among counties. In the absence of barriers regarding the movement of goods, services and production factors, certain countries and regions form powerful poles of attraction, leading to imbalances among countries with high differences in per capita income. The developed regions and countries become attraction poles that absorb more and more capital and qualified workforce from the less developed countries and regions. Therefore, well-off countries and regions score important economic growth and a more intense activity regarding the high-quality production factors whereas the poorly developed regions experience stagnation and economic downtrend. Under these circumstances, we can no longer speak of convergence but rather of divergence. Barro and Sala-i-Martin developed a series of convergence tests based on Solow’s growth model, in order to speed the catching-up process by less developed counties. They supported the neoclassical approach regarding the convergence in GDP per inhabitant growth between 1950 and 1985. Subsequently, Armstrong confirmed Barro and Sala-i-Martin’s conclusions as a result of his study that states the significant decrease in regional disparities (the convergence process stimulated by the migration of population) until the mid ‘70s, a period marked by the economic crisis (the oil crisis) and the process of economic divergence. [13] The empirical test of these hypotheses led to the quantitative definition of two types of real convergence: beta (β) and sigma (σ) convergence. Beta convergence (β) indicates that poorer economies have the long-term
tendency to evolve faster than richer ones in terms of absolute convergence hypothesis, whereas under conditional convergence hypothesis, the same phenomenon depends on a number of given factors. According to the neoclassical theory, poorer economies evolve faster than richer ones, which translates in the decrease of the dispersion coefficient of GDP per inhabitant, on the one hand, and the existence of an reverse, negative relation between the growth rate of GDP per inhabitant in a certain period of time and the initial level of GDP per inhabitant, on the other hand. The absolute β convergence hypothesis is usually tested by means of the relation:
ln(Yi ,T ) ln(Yi , 0 ) T
* ln(Yi ,0 ) i ,T
Where: Yi ,T - GDP per inhabitant in region i at moment t; T - The length of study period; , - Unknown parameters to be estimated econometrically; i,T - Random variables which reflect the error concerning the convergence level attained. β convergence is the situation when the coefficient is negative and statistically significant, which means the average growth rate of GDP per inhabitant is negatively correlated with the initial level of GDP per inhabitant. The beta coefficient analyzes the income mobility within the same dispersions studied by sigma convergence in terms of time evolution. The conditional β convergence hypothesis is usually tested by way of the following relation; the model estimations must be isolated and the variables that contribute to region differentiation kept constant:
ln(Yi ,T ) ln(Yi ,0 )
T
* ln(Yi ,0 ) X i i ,T
Where: Xi – variable vector which allows the constant maintenance of the stationary state of economy (the maintenance at a constant level of the economic balance). The beta parameter value indicates the intensity of the convergence process. This may have different values depending on the distance
731
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
between the economy in question and its long-term equilibrium. As economy draws closer to the state of equilibrium, the value of the coefficient diminishes. β convergence represents a necessary but insufficient condition for σ convergence, whereas internal disparities can diminish if poorer counties grow faster than richer ones. It is also possible for a country to grow much faster than others without reducing the development disparities. Sigma convergence (σ) indicates a descending variation of the level of income per inhabitant in a group of countries or regions, indicating the diminishing tendency of income distribution. It is used to describe the convergence level while analysing dispersion and its reduction measured by the standard deviation of the log of income or of the production per inhabitant (Barro and Sala-i-Martin). In the articles, they used the following formula so as to measure sigma convergence:
1 n y i log n i 1 y *
4. Testing beta and sigma convergence within development regions of Romania In our country, like in other new EU members, we can notice the existence of a centre-periphery structure which is selfsustained by the faster development of regions around capitals due to investments preferring these areas, workforce migration, and government aid in the developed regions to obtain a higher growth rate at national level. Obviously, the regional disparities among development regions of Romania continue to persist. To test the convergence hypothesis of the 8 development regions of Romania, we considered the econometric parameter β which indicates the convergence speed, while σ shows the convergence or divergence trend (the dimension of the data sample subject to analysis). The majority of recent studies point out that catching up with richer states or the so-called beta convergence is very rare. With a view to ensuring data comparability, GDP per inhabitant was calculated at comparable prices. The calculation of the regression equation was achieved in terms of the evolution of GDP per inhabitant in the development regions in 1998 (the first analysis year) and the average annual growth rates of GDP per inhabitant in the period between 1998 and 2006 (fig.1). The explanatory (independent) variable is GDP per inhabitant in the initial year, and the explained (dependent) variable, the average annual growth rate of GDP per inhabitant. The result of regression calculations was a positive β coefficient of the initial explanatory variable. This indicates the lack of convergence at development region level, which can also be noticed in figure 1. The result is due to discrepancies, to the large differences between the poorly developed regions and the developed ones with regard to the presence and the capacity of economic growth indicators (human and physical capital, technological progress) to produce higher economic growth, as well as the capacity of developed regions to absorb direct foreign investments, and to generate and assimilate new technologies [8].
2
The most frequently used is the variation coefficient: CvT
T
XT
,
Where: CvT = the variation coefficient in T period σT = mean square deviation of regional development in T period, calculated as:
T
1 n X iT X T n i 1
2
X T = average development level in T period. The sigma convergence test is based on the calculation of the disparities in GDP per inhabitant while the variation coefficient in T period is compared to T+1. Sigma convergence exists if CvT > CvT 1 , when the dispersion phenomenon is decreasing in a certain period of time.
732
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
Rata medie anula de crestere a PIB pe locuitor (1998-2006)
Figure 1. Average annual growth rate of GDP per inhabitant (1998 - 2006) and the initial level of regional development (1998) 14.00 12.00 10.00 8.00 6.00 4.00 2.00 0.00 0
2000
4000
6000
8000
10000
12000
14000
16000
PIB pe locuitor din anul 1998
Source: individual calculations based on the Romanian Statistical Yearbook of 2004 and 2008. The existence of inequalities at national level can be demonstrated by testing the real convergence process. The tests for the sigma convergence hypothesis applicable to the eight development regions of Romania was carried out for the period between 1998 and 2006 in order to determine the extent to which the regions have succeeded to reduce the differences among them throughout this
period of time. GDP per inhabitant at comparable prices, for comparability reasons, the data employed refer to the period between 1998 and 2006, based on the Romanian Statistical Yearbook of 2008, corrected with the GDP deflator; was used to calculate sigma convergence. The results obtained are shown in table 1.
Table 1 GDP per inhabitant variation coefficient (sigma convergence) for the development regions of Romania Years σT
XT CvT
1998
1999
2000
2001
2002
2003
2004
2005
2006
2084.197
2454.683
3564.608593
4014.648
4097.038
4548.125
5105.751
6528.957
7265.468
8899.16
9019.43
9174.92
9950.10
10840.60
12273.60
13782.67
14857.91
16680.00
0.23
0.27
0.39
0.40
0.38
0.37
0.37
0.44
0.44
Source: individual calculations based on the Romanian Statistical Yearbook of 2007, 2008; Variation coefficient - Graphical evolution (fig.2)
733
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
Coeficientul de variatie a PIB pe locuitor
Fig. 2 Graphical evolution of sigma convergence in development regions 0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
Ani
Source: chart based on data in Table 1. Y* = GDP per capita at purchasing parity level for the target country or region; g* = the growth rate of GDP per inhabitant for the target country/region. After determining the logarithms and rearranging the equation factors, we determine the period of time (t) necessary to achieve convergence between the two entities (the country/region for which convergence is studied and the target region/country) concerning the GDP per inhabitant:
By analysing the data in table 1 and the coefficient of variation of GDP per inhabitant (fig. 2), a growth is seen in the analysed period of time, which indicates a divergent economic growth trend. This type of convergence is not achieved in the regions of our country, as shown by the existence of the growth trend, while the differences among the incomes of regions tend to grow. Out of the eight development regions of Romania, Bucureşti Ilfov registers the highest GDP per inhabitant. The region will be used both as a basis for comparing the weights and as target level for revealing the necessary period of time for the other regions to achieve the regional convergence process. In order to establish the adequate period for a state or region to catch up with a target state or region in terms of living standard, we propose the following equation:
log Y * log Yr t log(1 g r ) log(1 g * ) The other development regions of Romania can catch up with Bucureşti Ilfov if their growth rates are higher than those of Bucureşti Ilfov. In order to calculate the period necessary for the development regions to reach the GDP per inhabitant in the Bucureşti Ilfov region, we consider an average annual growth rate of 3%, 5% respectively. The convergence speed of these regions in order to reach 100% of GDP per inhabitant in Bucureşti Ilfov is calculated in Table 2.
Yr 1 g r Y * (1 g * ) n n
Where: Yr represents GDP per capita at purchasing parity level for the country or region where convergence is studied (a less developed region). gr = the growth rate of GDP per inhabitant for the country/region where convergence is studied;
734
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
Table 2 Convergence between GDP per inhabitant of development regions of Romania and Bucureşti –Ilfov (Bucureşti – Ilfov growth rate of 3% and 5%) Weight of GDP per inhabitant in regions from GDP per inhabitant in Bucureşti Ilfov % (GDP per inhabitant in Bucureşti Ilfov =100 in 2006) 29.41 38.76 38.20
Average annual growth rate (19982006)
Years necessary to reach 100% of GDP per inhabitant in Bucureşti Ilfov (3%)
Years necessary to reach 100% of GDP per inhabitant in Bucureşti Ilfov (5%)
Region North East 5.05% 62 South East 5.69% 37 South Muntenia 7.56% 22 South-West Oltenia 35.60 5.97% 36 West 53.04 9.81% 10 North-West 42.69 7.61% 19 Centre 45.47 7.07% 20 Source: calculations based on the Romanian Statistical Yearbook of 2007, 2008; The average annual growth rate in the development regions was calculated for the period between 1998 and 2006, using the data provided by the Romanian Statistical Yearbook (table 2). The table shows that the regions with the lowest performances with respect to the convergence speed are North East, South East and South West Oltenia. If these regions start the convergence process with a growth of 5.05%, 5.69% and 5.97% respectively (growth registered in the period between 1998 and 2006) while Bucureşti Ilfov registers a constant growth rate of 3%, the GDP per inhabitant in the three regions will be equal to that of Bucureşti Ilfov in approximately 62 years (North East), 37 years (South-East) and 36 years (South-West Oltenia). Supposing that the average annual growth rate of GDP per inhabitant in Bucureşti Ilfov is 5%, the period of time necessary for the other regions to reach 100% of GDP per inhabitant in Bucureşti Ilfov is 2571 years (North East), 145 years (South East), 40 years (South Muntenia), 112 years (SouthWest Oltenia), 14 years (West), 35 years (North West) and 40 years Central region. The best performance is scored by the Western region; if its average annual growth rate continues to be 9.81%, it will reach the level of Bucureşti Ilfov within approximately 10 years (in the case of a growth rate of 3%)
2571 145 40 112 14 35 40
and 14 years respectively (in the case of a growth rate of 5%). The Western region would most rapidly catch up with Bucureşti Ilfov, followed by the North West region, while the North-East region would be the slowest in catching up with Bucureşti Ilfov. The next step consists in calculating the number of years necessary for the development regions of Romania to reach the EU-27 average (table 3). In order to see whether the development regions of Romania will succeed in achieving the convergence with EU-27 concerning the GDP per inhabitant, we should compare the development speed of our regions to that of the EU member states. [9] The average annual growth rate for all development regions and for the European Union was calculated based on the Eurostat data for the period between 2002 and 2006. Taking into account the the GDP per inhabitant and the current growth rates in the Romanian regions are quite low, it is very likely that a number of years will be necessary to reach a level of GDP per inhabitant comparable to that of the EU-27. 7 out of the 8 regions have a GDP per inhabitant 50% lower than the EU-27 average, which increases the catching-up period.
735
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
Table 3 Convergence in GDP per inhabitant of development regions of Romania with the EU-27 (for an EU-27 growth rate of 5.62%) Weight of GDP per inhabitant in the EU27 average at PPS(EURegion 27=100 in 2006) North East 24.7 South East 32.5 South Muntenia 32.1 South-West Oltenia 30.4 West 44.7 North-West 35.9 Centre 38.3 Bucureşti Ilfov 83.8 Source: calculations based on Eurostat data;
Average annual growth rate % (2002-2006)
Years necessary to reach 100% of the EU-27 average
10.03 9.63 7.15 9.28 11.6 13.10 11.25 12.57
97 33 21 23 13 25 26 3
According to the data in table 3, the Bucureşti Ilfov region, with a growth rate of 12.57%, would reach the EU-27 level in three years (for an EU average annual growth rate of 5.62%). The Western region would need 13 years to achieve the convergence. Conversely, the North Eastern region, despite the 10.03% growth rate of GDP per inhabitant, would need almost 97 years to attain convergence, due to the low GDP per inhabitant (the weight of the GDP per inhabitant as compared to the EU average is only 27.7% ).
Although statistical data fluctuates in time due to local factors, the geographical component has played a significant role in the economic growth. The underdeveloped regions are concentrated in the North-East, at the Romanian-Moldavian border and towards South, along the Danube. Underdevelopment seems to be largely correlated with unemployment and rural activity as well as with the incapacity to absorb direct foreign investments. Romanian economy presents some features that question its national regional policies and the problems that need to be solved, in its capacity as an economy which must integrate in the EU productive system, i.e.: unequal regional development within its borders; a distorted non-competitive sectorial structure; ageing, declining population; immigration of working-age population. The existence of regional inequalities is also reflected by testing the (beta and sigma) unconditional convergence hypotheses for the eight development regions of Romania. The results indicate the inexistence of the regional convergence process – the differences among regional income tends to grow – which means that certain regions represent strong attraction poles that absorb larger amounts of capital and high-quality workforce to the detriment of less developed regions. The regional development disparities may be due to various natural and human characteristics, as well as to relatively specific evolution frameworks (economic, technological, demographical, social, political and cultural) that have shaped their development in time. This has led to the predominance of agriculture
5. Conclusions Regional development represents the socioeconomic evolution of a certain region; its aims are to stimulate and diversify the economic activities, increase private sector investments and reduce unemployment in order to improve the standards of living. This can only be achieved by means of economic growth, focus on the individual resources and the role of regional socioeconomic development actors. In Romania, regional development appeared as a necessity to correct the existing regional disparities, on the one hand, and to apply the European Union legislation in this field, on the other hand. The evolution of imbalances is influenced by high specialization, which leads to an increased concentration level in certain areas which ensure the appropriate conditions and avoidance of other regions which are isolated or hardly accessible. A high level and dynamics of national activities tend to reduce the concentration level, while specialization and other sociological and cultural factors facilitate the differential distribution of activities.
736
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
as economic force in the regions with favourable weather conditions, of heavy industry in regions of iron and coal and of the tertiary sector branches in the administrative centres. The European Union sustains the catching-up process among the various areas of the member states so that all regions and their inhabitants may benefit from its social and economic advantages. This can be achieved by stimulating the elaboration of regional development strategies and the financial support of the development programs in the disadvantaged regions. Romania takes advantage of the cohesion policy and structural funds provided by the European Union. This opportunity, if used appropriately, may contribute to the economic development of our country and to the catching up with the other EU member sates. Regional disparities persist among development regions of Romania, even though low income regions benefit from a higher percentage of structural funds. In our country, like in other new EU member states, there is a centre-periphery structure that is self-sustained by the rapid growth of regions around the capital cities due to: investments which prefer developed regions, workforce immigration, government aids in poorly developed regions in order to obtain a higher national growth rate. Consequently, a regional policy designed to reduce regional disparities at national and European level must be promoted. Regional policy is determined by the national economic structure, the capitalization of the production and human resources potential. Demographic policies, education and employment policies are different according to the goals on the regional policy. Forming and developing an industrialagricultural structure based on the regional and national economic system may be adequate for Romania. Such a policy must be associated with incentive measures for birth rate, reducing the workforce migration, changing the educational structure, professional reconversion, developing managerial skills. At the same time, this type of approach implies a national regional policy as an alternative to the deindustrialization process and the stabilization of migrating population. All at once, material and financial resources must be used to create the necessary infrastructure. The consequences of such a regional policy may be: income and employment growth, increased demand for goods and
services, improved living standards, demographic dynamics. Regional development policies are associated with structural and sectorial adjustment policies, leading to improved competitiveness, which stimulates the complex process of catching-up of the regions lagging behind. References [1] Bal Ana, Luţaş Mihaela, Jora O., Topan V., 2007, Scenarii privind evoluţiile comunitare în domeniul competitivităţii, politicii de coeziune şi politicii de dezvoltare regională, Institutul European din România – Studii de strategie şi politici (SPOS 2007), Bucureşti; [2] Constantin Daniela-Luminiţa, 2004, Elemente fundamentale de economie regională., Editura ASE, Bucureşti; [3] Dinu M., Socol C., Marinaş M., 2005, Mecanisme de convergenţă şi coeziune, Editura Economică, Bucureşti; [4] Dinu M., Socol C., Niculescu Aura,2006, Fundamentarea şi coordonarea politicilor economice în Uniunea Europeană, Editura Economică, Bucureşti; [5] Dornbusch R., Fischer S., Startz R., Macroeconomie, Editura Economică, Bucureşti, 2007; [6] Egger P., 2005, Spatial beta and sigma convergence: Theoretical foundation, econometric estimation and an application to the growth of European region, Paper prepared for Spatial Econometrics Workshop, University of Kiel; [7] Frankel J.,2004, Real convergence and Euro Adoption in Central and Eastern Europe: Trade Gaubert N., La politique regionale europeenne entre convergence et cohesion, Colloque ASRDLF – Bruxelles ; [8] Iancu A.,2008, Convergenţa reală, Seria Working Papers, nr. 1, Academia Română, Institutul Naţional de Cercetări Economice, Programul CEEX, Proiectul: Convergenţa economică şi rolul cunoaşterii în condiţiile integrării în UE, Bucureşti; [9] Iancu Aurel, 2006, Problema convergenţei economice, Supliment al Revistei de Economie teoretică şi aplicată: România în Uniunea Europeană. Potenţialul de convergenţă, Bucureşti; [10] Manolescu G., Istudor N., 2007, Logistica proiectării dezvoltării regionale- Analiză, Strategie, Programare, Finanţare, Management, ASE, Bucuresti; [11] Marinaş M., 2006, Analiza corelaţiei dintre convergenţa nominală şi reală. Cazul României, revisa, Economie teoretică şi aplicată, nr. 3 (498), Editura AGER-Economistul, Bucureşti; [12] Marinaş M.,2008, Convergenţa economică, Editura Economică, Bucureşti; [13] Sala-i-Martin X., 1996, Regional Cohesion: Evidence and Theories of Regional Growth and Convergence, European Economic Review, no. 40.
737
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
CONVERGENCE IN THE DEVELOPMENT REGIONS OF ROMANIA Associate Professor Ph.D SÎRGHI Nicoleta West University of Timisoara,Faculty of Economics and Business Administration [email protected] Lecturer Ph.D PĂREAN Mihai West University of Timisoara, Faculty of Economics and Business Administration [email protected] Assistant Lecturer OŢIL Maria West University of Timisoara, Faculty of Economics and Business Administration [email protected]
Abstract
The present paper consists of two parts. Part one focuses on regional growth theories which have registered a gradual evolution, marked, in the beginning, by the analysis of regions as non-spatial elements of national economy. Part two suggests an econometric testing of the two types of real convergence: beta convergence and sigma convergence in the development regions of Romania. Thus, the period of time required by the development regions of Romania to reduce the discrepancies in GDP per inhabitant compared to Bucureşti Ilfov region as well as to the EU-27 average is determined.
Convergence is a dynamic process based on the application of socio-economic policies designed to reduce disparities among regions and countries in a given space. It is achieved by implementing the structural policy in order to obtain sharp economic growth parameters in the peripheral regions which experienced a period of economic downtrend or didn’t manage to attain the economic performance of the region to which they belong. Regional growth is considered of utmost importance when referring to the real economic situation of a development region.
the comparative costs, etc.) which determine how the given economic growth rate will be distributed among the system’s regions. Within these models, a region’s growth will always bring about the other’s downfall. Many of the traditional regional growth theories (cumulative causality, neoclassical theory, etc.) consider regional growth as competitive growth. The national growth rate is determined exogenously, while regional economic analysis is concerned only with the distribution of growth among regions. The need to include in these variable spatial models is reduced since every region is considered an economic sector. However, according to the generative growth models, the regional dimension is far more accentuated, as these consider national economic growth rate as the resultant of regional growth rates. Consequently, the entire growth is spatially oriented, in other words the growth of any national economic
Key words: convergence, development, sigma convergence (σ), beta convergence (β). 1. Introduction The concept of regional growth has evolved gradually in time and was marked in the beginning by the analysis of regions as non-spatial elements of the national economy. One of the consequences is that regional growth rates are considered to derive from the national growth rates, the propulsive impact of regional growth on the national growth rate being ignored. This way of approaching regional growth emphasizes the idea of interregional competition, hence the competitive growth concept. The competitive growth models imply that the attainable rate of national growth is known, thus examining the forces (the advantages and disadvantages of location, the relative potential of the market,
727
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
sector has its origins in a certain location. Regional growth performances can be improved without producing adverse effects on the growth rate of the neighbouring regions. Growth generated by the innovation process can be integrated into the following context: the agglomerations and the spatial proximity of activities in certain towns or regions may lead to an innovation rate above that registered in the absence of agglomerations. Similarly, the changes manifested within the intraregional distribution of productive inputs, enabled by an efficient intraregional transport system for example, may increase both production efficiency and regional growth rates. The importance of spatial impact on regional growth is overlooked in the case of competitive growth models. The phenomenon that allows intraregional spatial efficiency to have a feedback effect on the aggregate growth rate is called generative growth. These brief considerations reinforce the necessity of taking into account the spatial component when elaborating the regional economic growth theory and the urban-regional analysis. In practice, it is reflected in the careful study of the implications generated by the adoption of certain regional development policies which differ depending on the amplitude of political power interventions, the centralization or decentralization of regional policy implementation, the macroeconomic or microeconomic orientation, forecast measures having an inter and intraregional field of action, the focus on the redistribution of workforce or capital, the exogenous sources or the endogenous development potential, the upgrading process, the expansion of the tertiary sector, or the traditional sectorial measures. At the same time, a region can present a reduced overall output growth and a rapid increase of the output per capita if the immigration process in that region registers significant quotas during the study period. The most commonly encountered are: a region’s overall output growth, growth of the output per occupied person and per capita. In its turn, the output can be determined by dint of: total (gross) output of a region, the regional GDP, the regional net domestic product, etc. The problem is finding the most adequate way to measure regional growth. This also depends on the purpose of
the measure. Therefore, overall output growth is used to indicate the growth of the regional production capacity, which partially depends on the extent to which the region attracts capital and labour force from other regions. The growth of the output per occupied person is often used as an indicator for the modifications in regional competitiveness by means of productivity growth. The growth of output per capita indicates changes in the economic prosperity at region level. Another controversial issue in relation to the regional growth disparities refers to the medium and long-term vision concerning the regional growth consequences. Thus, according to the neoclassical model, which highlights the role of offer, regional growth leads to convergence in the regional socioeconomic development, while conforming to the models based on the postKeynesian approach to demand (the model based on the export potential, the cumulative model), regional growth underlines divergence. Their complementary approach was outlined against the two visions on regional growth consequences, depending on the general level of development, the economic policy options of a country and issues specific to different regions of the territorial system. 2. Literature review The liberal doctrine of development is based on three major concepts: the economic growth pole, the industrial complex and the industrial district. The growth pole concept (growth pole strategy) was developed by Francois Perroux. Since growth pole level is accompanied by a superior innovative and competitive capacity, the advanced areas may generate a continuously polarized development. In the absence of barriers concerning the workforce and population migration, the appearance of growth poles may be associated with important population displacements which represent another demographic issue[1]. Generally, the direction of these displacements is from the insufficiently developed regions towards developed ones and concerns the young and middle-aged population as well as university graduates. As a result, the developed regions impose a series of filtering processes on the input flows of workforce. Such phenomena lead to a mainly qualitative depopulation of
728
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
the insufficiently developed countries, influence the rhythm of demographic evolution (marriage and birth rate), emphasizing demographic ageing and producing a decrease in regional competitiveness. The developed regions’ filtering of activities and the competitiveness loss of the less developed regions, together with the qualitative depopulation may lead to economic duality. According to liberals, the main causes of regional development disparities are: the lack of entrepreneurship education and excessive state interventions. The revival of the insufficiently developed regions can be achieved by means of labour market deregulation, free functioning of markets, and privatization of public enterprises. According to the Keynesian interventionist approach, the main causes of regional disparities are market insufficiencies and capital concentration. The solution for reducing disparities would be the intervention on behalf of public, national, regional and local authorities by means of public investments into the socioeconomic infrastructure [2]. The incentive practice may be a way towards regional development. The main incentives are: tax decrease; land lease; subsidized credits; employment growth incentives; facilities for SMEs. A strongly controlled economic policy may prevent the territorial and economic disparities in the short run, leading instead to mainly political artificial privileges, to the disadvantage of public and private sector. This causes hidden economy development, enlarged deficit, public debt growth, disinvestment, increase in inflation and decrease in income growth and tax basis. In conclusion, interventionism would produce stagnation and long-term increase in inequalities, i.e. exactly the opposite of initial intentions, thus imposing financial liberalization measures and restrictive monetary policies. At the beginning of the ‘70s, regional development mainly focused on the eradication of poverty and development of small towns. Thus the idea of development by means of central growth was abandoned. The political problems at that time indicate that the national growth objectives (regional) differed from the rural development ones in terms of anticipatory effects. Therefore, the necessity of studying the implications of
rural development policy based on the growth poles was acknowledged. Subsequently, other policies and concepts were developed referring to the urban centre development in conjunction with rural development. Initially, it was the states that undertook to reduce regional disparities. Within a short period of time, the impact of implementing regional development programs achieved by the state and mainly based on fiscal incentives proved to be quite ineffective [3]. The revival of regionalist movements in the ‘70s, the implementation of administrative reforms in a number of countries that passed laws related to regional level, the European Community enlargement and the increasing transfer of duties towards community institutions contributed to the radical change in the regional development phenomenon. Now, regional development has different meanings in the European countries. Within the framework of the European Union, few states analyse regional development in terms of the major problems posed by economic development of internal territories compared to the economically developed and privileged regions from the geographical, historical and resource endowments point of view, etc. Others consider the regional development approach to ensure the effective and active participation of all citizens to the economic, political and social development of the country. Thirdly, it implies the development of the territories far from the economically developed centre. However, in all the abovementioned situations, regional development is an already traditional and consistently promoted government policy sustained by community funds. In Central and Eastern European countries, regional development is a relatively recent phenomenon, intensely promoted by dint of the democratization process and the intensification of socioeconomic development while maintaining and supporting the territorial integrity and expanding the interstate collaboration. The focus is on the creation of units, the so-called territorial regions, distinct from the existing intern administrative-territorial state structure and on the general balanced socioeconomic growth of these structures. In a short period of time new priorities have been established
729
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
for territorial development, objectives have been modified, an institutional framework for the financing of development programs was created and the instruments for regional development policy implementation diversified so as to concretely underline the development level and to offer an overall report of territorial units for regional statistic accomplishment thanks to the implementation of the Nomenclature of Territorial Units for Statistics (NUTS).
takes place when the economic growth gap is widening. GDP per inhabitant at purchasing power parity represents the classic real convergence indicator [4] Solow’s economic growth model demonstrates the possibility of catching up with richer countries as far as income per inhabitant is concerned. The neoclassical theory based on this model upholds the possibility of achieving the economic convergence process among countries. The fundamental hypothesis of the neoclassical model is "the dependence of convergence (catching-up process) on the return on capital, on its generally decreasing trend. Capital growth will correspond to a less than proportional return on capital increase.” [9]. The poor countries, with a low GDP and physical capital per inhabitant, grow faster compared to richer economies with a higher GDP and capital per inhabitant. The neoclassical economic growth theory indicates that all economies characterized by the same basic parameters will attain an equal level of development, regardless of their initial state. As a result, real convergence hypotheses were formulated based on the convergence theory conclusions: absolute and conditional convergence hypotheses. According to the absolute convergence hypothesis, the farther a country is from its long-term equilibrium, the higher the income per inhabitant; the initially poor counties will evolve faster than the initially richer ones. Absolute economic convergence appears when there is a reverse relation between the initial level of income and the economic growth rate, i.e. ”the economic growth rate decreases as the income level reaches the steady state”[10] . The conditional convergence hypothesis implies that if certain structural parameters differ, i.e. the analysed countries or regions do not possess the same structural characteristics, then „each of them will converge towards its own limits” [1]. Although growth rates may be higher in developed countries than in richer countries, the development differences remain significant (they are maintained and even accentuated, leading to income divergence). This convergence type implies that all economies with different initial capital stock per capita have the same savings rate, similar technologies, as well as the same population growth rates. In case of non-
3. Hypotheses and convergence types Convergence represents a dynamic process based on the application of socioeconomic policies designed to reduce disparities among regions and countries in a given space. It is achieved by means of structural policy implementation in view of obtaining sharp economic growth parameters in the peripheral regions which experienced a period of economic downtrend or didn’t attain the economic performance of the region to which they belong. Another definition of convergence pertains to the increase in similarities and economic performance of regional and national economies in a given space. Convergence is seen as a preliminary condition for integration because while policy implementation and creation structures converge, the integration process together with its strategies and the establishment of common functional institutions are more easily achieved. Therefore, convergence means working towards the same goal, evolving in order to reach a target. The convergence problem lies in “establishing the extent to which economies with different initial output levels register a growth that allows them to finally reach equal living standards” [1]. The convergence process has in view the approximation of the economic, social, monetary, fiscal and performance indicators of regions and countries, the reduction of development disparities, the assurance of financial and monetary stability in all the countries as well as the compatibility of institutional and regional administrative structures and mechanisms. We can speak about real economic convergence if poor countries’ economic growth rate surpasses that of the richer states, reducing income disparities among them. The economic divergence process
730
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
fulfilment of conditions, convergence cannot occur. Later, Mydral (1957) replaces the stable equilibrium hypothesis with the circular cumulative causation which determines the divergence phenomenon [13]. The hypothesis explains the persistence and widening of international and interregional differences at development level. Regional and international inequalities in different countries are perpetuated and even widened by means of capital movement, human capital and workforce migration, and the exchanges of goods and services. It’s the polarisation trend (agglomeration) in the conditions of economic and monetary integration that accentuates the level of disparity among counties. In the absence of barriers regarding the movement of goods, services and production factors, certain countries and regions form powerful poles of attraction, leading to imbalances among countries with high differences in per capita income. The developed regions and countries become attraction poles that absorb more and more capital and qualified workforce from the less developed countries and regions. Therefore, well-off countries and regions score important economic growth and a more intense activity regarding the high-quality production factors whereas the poorly developed regions experience stagnation and economic downtrend. Under these circumstances, we can no longer speak of convergence but rather of divergence. Barro and Sala-i-Martin developed a series of convergence tests based on Solow’s growth model, in order to speed the catching-up process by less developed counties. They supported the neoclassical approach regarding the convergence in GDP per inhabitant growth between 1950 and 1985. Subsequently, Armstrong confirmed Barro and Sala-i-Martin’s conclusions as a result of his study that states the significant decrease in regional disparities (the convergence process stimulated by the migration of population) until the mid ‘70s, a period marked by the economic crisis (the oil crisis) and the process of economic divergence. [13] The empirical test of these hypotheses led to the quantitative definition of two types of real convergence: beta (β) and sigma (σ) convergence. Beta convergence (β) indicates that poorer economies have the long-term
tendency to evolve faster than richer ones in terms of absolute convergence hypothesis, whereas under conditional convergence hypothesis, the same phenomenon depends on a number of given factors. According to the neoclassical theory, poorer economies evolve faster than richer ones, which translates in the decrease of the dispersion coefficient of GDP per inhabitant, on the one hand, and the existence of an reverse, negative relation between the growth rate of GDP per inhabitant in a certain period of time and the initial level of GDP per inhabitant, on the other hand. The absolute β convergence hypothesis is usually tested by means of the relation:
ln(Yi ,T ) ln(Yi , 0 ) T
* ln(Yi ,0 ) i ,T
Where: Yi ,T - GDP per inhabitant in region i at moment t; T - The length of study period; , - Unknown parameters to be estimated econometrically; i,T - Random variables which reflect the error concerning the convergence level attained. β convergence is the situation when the coefficient is negative and statistically significant, which means the average growth rate of GDP per inhabitant is negatively correlated with the initial level of GDP per inhabitant. The beta coefficient analyzes the income mobility within the same dispersions studied by sigma convergence in terms of time evolution. The conditional β convergence hypothesis is usually tested by way of the following relation; the model estimations must be isolated and the variables that contribute to region differentiation kept constant:
ln(Yi ,T ) ln(Yi ,0 )
T
* ln(Yi ,0 ) X i i ,T
Where: Xi – variable vector which allows the constant maintenance of the stationary state of economy (the maintenance at a constant level of the economic balance). The beta parameter value indicates the intensity of the convergence process. This may have different values depending on the distance
731
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
between the economy in question and its long-term equilibrium. As economy draws closer to the state of equilibrium, the value of the coefficient diminishes. β convergence represents a necessary but insufficient condition for σ convergence, whereas internal disparities can diminish if poorer counties grow faster than richer ones. It is also possible for a country to grow much faster than others without reducing the development disparities. Sigma convergence (σ) indicates a descending variation of the level of income per inhabitant in a group of countries or regions, indicating the diminishing tendency of income distribution. It is used to describe the convergence level while analysing dispersion and its reduction measured by the standard deviation of the log of income or of the production per inhabitant (Barro and Sala-i-Martin). In the articles, they used the following formula so as to measure sigma convergence:
1 n y i log n i 1 y *
4. Testing beta and sigma convergence within development regions of Romania In our country, like in other new EU members, we can notice the existence of a centre-periphery structure which is selfsustained by the faster development of regions around capitals due to investments preferring these areas, workforce migration, and government aid in the developed regions to obtain a higher growth rate at national level. Obviously, the regional disparities among development regions of Romania continue to persist. To test the convergence hypothesis of the 8 development regions of Romania, we considered the econometric parameter β which indicates the convergence speed, while σ shows the convergence or divergence trend (the dimension of the data sample subject to analysis). The majority of recent studies point out that catching up with richer states or the so-called beta convergence is very rare. With a view to ensuring data comparability, GDP per inhabitant was calculated at comparable prices. The calculation of the regression equation was achieved in terms of the evolution of GDP per inhabitant in the development regions in 1998 (the first analysis year) and the average annual growth rates of GDP per inhabitant in the period between 1998 and 2006 (fig.1). The explanatory (independent) variable is GDP per inhabitant in the initial year, and the explained (dependent) variable, the average annual growth rate of GDP per inhabitant. The result of regression calculations was a positive β coefficient of the initial explanatory variable. This indicates the lack of convergence at development region level, which can also be noticed in figure 1. The result is due to discrepancies, to the large differences between the poorly developed regions and the developed ones with regard to the presence and the capacity of economic growth indicators (human and physical capital, technological progress) to produce higher economic growth, as well as the capacity of developed regions to absorb direct foreign investments, and to generate and assimilate new technologies [8].
2
The most frequently used is the variation coefficient: CvT
T
XT
,
Where: CvT = the variation coefficient in T period σT = mean square deviation of regional development in T period, calculated as:
T
1 n X iT X T n i 1
2
X T = average development level in T period. The sigma convergence test is based on the calculation of the disparities in GDP per inhabitant while the variation coefficient in T period is compared to T+1. Sigma convergence exists if CvT > CvT 1 , when the dispersion phenomenon is decreasing in a certain period of time.
732
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
Rata medie anula de crestere a PIB pe locuitor (1998-2006)
Figure 1. Average annual growth rate of GDP per inhabitant (1998 - 2006) and the initial level of regional development (1998) 14.00 12.00 10.00 8.00 6.00 4.00 2.00 0.00 0
2000
4000
6000
8000
10000
12000
14000
16000
PIB pe locuitor din anul 1998
Source: individual calculations based on the Romanian Statistical Yearbook of 2004 and 2008. The existence of inequalities at national level can be demonstrated by testing the real convergence process. The tests for the sigma convergence hypothesis applicable to the eight development regions of Romania was carried out for the period between 1998 and 2006 in order to determine the extent to which the regions have succeeded to reduce the differences among them throughout this
period of time. GDP per inhabitant at comparable prices, for comparability reasons, the data employed refer to the period between 1998 and 2006, based on the Romanian Statistical Yearbook of 2008, corrected with the GDP deflator; was used to calculate sigma convergence. The results obtained are shown in table 1.
Table 1 GDP per inhabitant variation coefficient (sigma convergence) for the development regions of Romania Years σT
XT CvT
1998
1999
2000
2001
2002
2003
2004
2005
2006
2084.197
2454.683
3564.608593
4014.648
4097.038
4548.125
5105.751
6528.957
7265.468
8899.16
9019.43
9174.92
9950.10
10840.60
12273.60
13782.67
14857.91
16680.00
0.23
0.27
0.39
0.40
0.38
0.37
0.37
0.44
0.44
Source: individual calculations based on the Romanian Statistical Yearbook of 2007, 2008; Variation coefficient - Graphical evolution (fig.2)
733
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
Coeficientul de variatie a PIB pe locuitor
Fig. 2 Graphical evolution of sigma convergence in development regions 0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
Ani
Source: chart based on data in Table 1. Y* = GDP per capita at purchasing parity level for the target country or region; g* = the growth rate of GDP per inhabitant for the target country/region. After determining the logarithms and rearranging the equation factors, we determine the period of time (t) necessary to achieve convergence between the two entities (the country/region for which convergence is studied and the target region/country) concerning the GDP per inhabitant:
By analysing the data in table 1 and the coefficient of variation of GDP per inhabitant (fig. 2), a growth is seen in the analysed period of time, which indicates a divergent economic growth trend. This type of convergence is not achieved in the regions of our country, as shown by the existence of the growth trend, while the differences among the incomes of regions tend to grow. Out of the eight development regions of Romania, Bucureşti Ilfov registers the highest GDP per inhabitant. The region will be used both as a basis for comparing the weights and as target level for revealing the necessary period of time for the other regions to achieve the regional convergence process. In order to establish the adequate period for a state or region to catch up with a target state or region in terms of living standard, we propose the following equation:
log Y * log Yr t log(1 g r ) log(1 g * ) The other development regions of Romania can catch up with Bucureşti Ilfov if their growth rates are higher than those of Bucureşti Ilfov. In order to calculate the period necessary for the development regions to reach the GDP per inhabitant in the Bucureşti Ilfov region, we consider an average annual growth rate of 3%, 5% respectively. The convergence speed of these regions in order to reach 100% of GDP per inhabitant in Bucureşti Ilfov is calculated in Table 2.
Yr 1 g r Y * (1 g * ) n n
Where: Yr represents GDP per capita at purchasing parity level for the country or region where convergence is studied (a less developed region). gr = the growth rate of GDP per inhabitant for the country/region where convergence is studied;
734
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
Table 2 Convergence between GDP per inhabitant of development regions of Romania and Bucureşti –Ilfov (Bucureşti – Ilfov growth rate of 3% and 5%) Weight of GDP per inhabitant in regions from GDP per inhabitant in Bucureşti Ilfov % (GDP per inhabitant in Bucureşti Ilfov =100 in 2006) 29.41 38.76 38.20
Average annual growth rate (19982006)
Years necessary to reach 100% of GDP per inhabitant in Bucureşti Ilfov (3%)
Years necessary to reach 100% of GDP per inhabitant in Bucureşti Ilfov (5%)
Region North East 5.05% 62 South East 5.69% 37 South Muntenia 7.56% 22 South-West Oltenia 35.60 5.97% 36 West 53.04 9.81% 10 North-West 42.69 7.61% 19 Centre 45.47 7.07% 20 Source: calculations based on the Romanian Statistical Yearbook of 2007, 2008; The average annual growth rate in the development regions was calculated for the period between 1998 and 2006, using the data provided by the Romanian Statistical Yearbook (table 2). The table shows that the regions with the lowest performances with respect to the convergence speed are North East, South East and South West Oltenia. If these regions start the convergence process with a growth of 5.05%, 5.69% and 5.97% respectively (growth registered in the period between 1998 and 2006) while Bucureşti Ilfov registers a constant growth rate of 3%, the GDP per inhabitant in the three regions will be equal to that of Bucureşti Ilfov in approximately 62 years (North East), 37 years (South-East) and 36 years (South-West Oltenia). Supposing that the average annual growth rate of GDP per inhabitant in Bucureşti Ilfov is 5%, the period of time necessary for the other regions to reach 100% of GDP per inhabitant in Bucureşti Ilfov is 2571 years (North East), 145 years (South East), 40 years (South Muntenia), 112 years (SouthWest Oltenia), 14 years (West), 35 years (North West) and 40 years Central region. The best performance is scored by the Western region; if its average annual growth rate continues to be 9.81%, it will reach the level of Bucureşti Ilfov within approximately 10 years (in the case of a growth rate of 3%)
2571 145 40 112 14 35 40
and 14 years respectively (in the case of a growth rate of 5%). The Western region would most rapidly catch up with Bucureşti Ilfov, followed by the North West region, while the North-East region would be the slowest in catching up with Bucureşti Ilfov. The next step consists in calculating the number of years necessary for the development regions of Romania to reach the EU-27 average (table 3). In order to see whether the development regions of Romania will succeed in achieving the convergence with EU-27 concerning the GDP per inhabitant, we should compare the development speed of our regions to that of the EU member states. [9] The average annual growth rate for all development regions and for the European Union was calculated based on the Eurostat data for the period between 2002 and 2006. Taking into account the the GDP per inhabitant and the current growth rates in the Romanian regions are quite low, it is very likely that a number of years will be necessary to reach a level of GDP per inhabitant comparable to that of the EU-27. 7 out of the 8 regions have a GDP per inhabitant 50% lower than the EU-27 average, which increases the catching-up period.
735
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
Table 3 Convergence in GDP per inhabitant of development regions of Romania with the EU-27 (for an EU-27 growth rate of 5.62%) Weight of GDP per inhabitant in the EU27 average at PPS(EURegion 27=100 in 2006) North East 24.7 South East 32.5 South Muntenia 32.1 South-West Oltenia 30.4 West 44.7 North-West 35.9 Centre 38.3 Bucureşti Ilfov 83.8 Source: calculations based on Eurostat data;
Average annual growth rate % (2002-2006)
Years necessary to reach 100% of the EU-27 average
10.03 9.63 7.15 9.28 11.6 13.10 11.25 12.57
97 33 21 23 13 25 26 3
According to the data in table 3, the Bucureşti Ilfov region, with a growth rate of 12.57%, would reach the EU-27 level in three years (for an EU average annual growth rate of 5.62%). The Western region would need 13 years to achieve the convergence. Conversely, the North Eastern region, despite the 10.03% growth rate of GDP per inhabitant, would need almost 97 years to attain convergence, due to the low GDP per inhabitant (the weight of the GDP per inhabitant as compared to the EU average is only 27.7% ).
Although statistical data fluctuates in time due to local factors, the geographical component has played a significant role in the economic growth. The underdeveloped regions are concentrated in the North-East, at the Romanian-Moldavian border and towards South, along the Danube. Underdevelopment seems to be largely correlated with unemployment and rural activity as well as with the incapacity to absorb direct foreign investments. Romanian economy presents some features that question its national regional policies and the problems that need to be solved, in its capacity as an economy which must integrate in the EU productive system, i.e.: unequal regional development within its borders; a distorted non-competitive sectorial structure; ageing, declining population; immigration of working-age population. The existence of regional inequalities is also reflected by testing the (beta and sigma) unconditional convergence hypotheses for the eight development regions of Romania. The results indicate the inexistence of the regional convergence process – the differences among regional income tends to grow – which means that certain regions represent strong attraction poles that absorb larger amounts of capital and high-quality workforce to the detriment of less developed regions. The regional development disparities may be due to various natural and human characteristics, as well as to relatively specific evolution frameworks (economic, technological, demographical, social, political and cultural) that have shaped their development in time. This has led to the predominance of agriculture
5. Conclusions Regional development represents the socioeconomic evolution of a certain region; its aims are to stimulate and diversify the economic activities, increase private sector investments and reduce unemployment in order to improve the standards of living. This can only be achieved by means of economic growth, focus on the individual resources and the role of regional socioeconomic development actors. In Romania, regional development appeared as a necessity to correct the existing regional disparities, on the one hand, and to apply the European Union legislation in this field, on the other hand. The evolution of imbalances is influenced by high specialization, which leads to an increased concentration level in certain areas which ensure the appropriate conditions and avoidance of other regions which are isolated or hardly accessible. A high level and dynamics of national activities tend to reduce the concentration level, while specialization and other sociological and cultural factors facilitate the differential distribution of activities.
736
The International Conference on Administration and Business
ICEA - FAA 2009
14 – 15 NOVEMBER 2009
http://conference.faa.ro
The Faculty of Business and Administration University of Bucharest
as economic force in the regions with favourable weather conditions, of heavy industry in regions of iron and coal and of the tertiary sector branches in the administrative centres. The European Union sustains the catching-up process among the various areas of the member states so that all regions and their inhabitants may benefit from its social and economic advantages. This can be achieved by stimulating the elaboration of regional development strategies and the financial support of the development programs in the disadvantaged regions. Romania takes advantage of the cohesion policy and structural funds provided by the European Union. This opportunity, if used appropriately, may contribute to the economic development of our country and to the catching up with the other EU member sates. Regional disparities persist among development regions of Romania, even though low income regions benefit from a higher percentage of structural funds. In our country, like in other new EU member states, there is a centre-periphery structure that is self-sustained by the rapid growth of regions around the capital cities due to: investments which prefer developed regions, workforce immigration, government aids in poorly developed regions in order to obtain a higher national growth rate. Consequently, a regional policy designed to reduce regional disparities at national and European level must be promoted. Regional policy is determined by the national economic structure, the capitalization of the production and human resources potential. Demographic policies, education and employment policies are different according to the goals on the regional policy. Forming and developing an industrialagricultural structure based on the regional and national economic system may be adequate for Romania. Such a policy must be associated with incentive measures for birth rate, reducing the workforce migration, changing the educational structure, professional reconversion, developing managerial skills. At the same time, this type of approach implies a national regional policy as an alternative to the deindustrialization process and the stabilization of migrating population. All at once, material and financial resources must be used to create the necessary infrastructure. The consequences of such a regional policy may be: income and employment growth, increased demand for goods and
services, improved living standards, demographic dynamics. Regional development policies are associated with structural and sectorial adjustment policies, leading to improved competitiveness, which stimulates the complex process of catching-up of the regions lagging behind. References [1] Bal Ana, Luţaş Mihaela, Jora O., Topan V., 2007, Scenarii privind evoluţiile comunitare în domeniul competitivităţii, politicii de coeziune şi politicii de dezvoltare regională, Institutul European din România – Studii de strategie şi politici (SPOS 2007), Bucureşti; [2] Constantin Daniela-Luminiţa, 2004, Elemente fundamentale de economie regională., Editura ASE, Bucureşti; [3] Dinu M., Socol C., Marinaş M., 2005, Mecanisme de convergenţă şi coeziune, Editura Economică, Bucureşti; [4] Dinu M., Socol C., Niculescu Aura,2006, Fundamentarea şi coordonarea politicilor economice în Uniunea Europeană, Editura Economică, Bucureşti; [5] Dornbusch R., Fischer S., Startz R., Macroeconomie, Editura Economică, Bucureşti, 2007; [6] Egger P., 2005, Spatial beta and sigma convergence: Theoretical foundation, econometric estimation and an application to the growth of European region, Paper prepared for Spatial Econometrics Workshop, University of Kiel; [7] Frankel J.,2004, Real convergence and Euro Adoption in Central and Eastern Europe: Trade Gaubert N., La politique regionale europeenne entre convergence et cohesion, Colloque ASRDLF – Bruxelles ; [8] Iancu A.,2008, Convergenţa reală, Seria Working Papers, nr. 1, Academia Română, Institutul Naţional de Cercetări Economice, Programul CEEX, Proiectul: Convergenţa economică şi rolul cunoaşterii în condiţiile integrării în UE, Bucureşti; [9] Iancu Aurel, 2006, Problema convergenţei economice, Supliment al Revistei de Economie teoretică şi aplicată: România în Uniunea Europeană. Potenţialul de convergenţă, Bucureşti; [10] Manolescu G., Istudor N., 2007, Logistica proiectării dezvoltării regionale- Analiză, Strategie, Programare, Finanţare, Management, ASE, Bucuresti; [11] Marinaş M., 2006, Analiza corelaţiei dintre convergenţa nominală şi reală. Cazul României, revisa, Economie teoretică şi aplicată, nr. 3 (498), Editura AGER-Economistul, Bucureşti; [12] Marinaş M.,2008, Convergenţa economică, Editura Economică, Bucureşti; [13] Sala-i-Martin X., 1996, Regional Cohesion: Evidence and Theories of Regional Growth and Convergence, European Economic Review, no. 40.
737
