40TH CONGRESS OF THE
EUROPEAN REGIONAL SCIENCE ASSOCIATION
29 AUGUST-1 SEPTEMBER 2000 IN BARCELONA
THE ROMANIAN REGIONS COMPETITIVENESS
Dorin JULA,
Nicoleta JULA,
Bucharest Ecological University, and
Politehnica University of Bucharest
Institute for Economic Forecast
Bucharest, Romania
Bucharest, Romania
njula@[Link]
djula@[Link]
Abstract
Over the latest years, the issue of competitiveness has been the subject a lot of studies.
Even if many of them have tended to concentrate on countries or industries, there are
remarkable studies concerning the regions competitiveness (for example, 6th Periodic
Report of the EU regions). In this paper we are providing an extension to the Romanian
regions of the European Commission analysis focused on EU region competitiveness.
Certainly, because of the lack of reliable data on the different aspects that determine
overall competitiveness, any analysis can only be partial. However, as in the
Commission studies, in Romania four factors emerged as being closely linked with
regional differences in the competitiveness: the structure of economic activity (the
division of employment between agriculture, manufacturing, construction and
services), the extent of innovative activity (employees from research activity), regional
accessibility (transport infrastructure) and the skills of the work force (training level).
1. DEFINING REGIONAL COMPETITIVENESS
The concept of competitiveness is relatively clear when applied to enterprises, but is
more difficult to define and measure the competitiveness when applied to regions or
countries. At microeconomic level, competitiveness is viewed as the " ability to
compete in markets for goods or services. This is based on a combination of price and
quality. With equal quality and an established reputation, suppliers are competitive only
if their prices are as low as those of rivals. A new supplier without an established
reputation may need a lower price than rivals to compete. With lower quality than
rivals, a firm may not be competitive even with a low price; with a reputation for
superior quality, a supplier may be competitive even with a higher price than rivals"
(Black J., 1997)
The concept of competitiveness is, unfortunately, more ambiguous when applied to
countries or regions. A problem in accordance with Krugman (1994) is that the
economic problem facing any modern nation is not essentially one of competing on
world markets which the United States and Japan are not competitors in the same
sense that Coca-Cola competes with Pepsi. "The idea that a country's economic
fortunes are largely determined by its success on world markets is a hypothesis, not a
necessary truth; and as a practical, empirical matter, that hypothesis is flatly wrong",
says Krugman. Another problem is that the term competitiveness tends to create the
impression of a zero-sum game, a win/lose situation, in which countries or regions can
improve their position only at the expense of others, whereas, in practice, there are
mutual gains to be achieved from individual regions becoming more competitive. And
there are many other questions in economic literature regarding regional or national
competitiveness.
In these conditions, the challenge is to develop a concept of competitiveness that avoids
these problems. The Institute for Management Development of Lausanne (authors of
the World Competitiveness Yearbook) define national competitiveness as: "the ability
of a country to create added value and thus increase national wealth by managing assets
and processes, attractiveness and aggressiveness, globality and proximity, and by
integrating these relationships into an economic and social model" (Institute of
Management Development, 1996).
2
Likewise, in an OECD study, competitiveness is defined as the "ability to produce
goods and services which meet the test of international markets, while at the same time
maintaining high and sustainable levels of income" or, more generally, "the ability of
companies, industries, regions, nations and supra-national regions to generate, while
being exposed to international competition, relatively high income and employment
levels"(Commission europenne, 1999).
At regional level, "it needs to capture the notion that, despite the fact that there are
strongly competitive and uncompetitive firms in every region, there are common
features within a region which affect the competitiveness of all firms located there.
These features include physical and social infrastructure, the skills of the work force
and the efficiency of public institutions. In an increasingly global economy, such
factors can contribute strongly to business success and need to be at least of a minimum
standard in order to avoid putting firms at a significant disadvantage as compared with
those located elsewhere" (Commission europenne, 1999).
Economic literature point out different ways in which academics and policymakers
currently think of national or regional competitiveness. However, this paper is not
aiming at an overview of economic literature concerning the concept of regional
competitiveness, but only at analysing the way that model developed by European
Union for the European regions can be applied to Romanian regions. In these
circumstances, I am using a regional competitiveness definition pointed out in the 6-th
Periodic Report on Regions (Commission europenne, 1999). In the beginning, I am
presenting a short economic analysis of Romanian regions. Afterwards the analysis is
focusing on the regional competitiveness elements.
2. ROMANIAN REGIONAL ECONOMY
2.1. Romanian Development Regions
Compared to the other European states, Romania is a middle-sized country, with a
territory of 238391 km2, and a population of 22455.5 thousands inhabitants (at January
1st 2000).
3
The territorial-administrative structure of Romania includes 263 towns (of which 84
municipalities) and 2688 communes (over 13 thousand villages are grouped in these
communes). The Romanian territory is organized in 41 counties (judets) plus
Bucharest, the country capital. In accordance with the Law 151/1998 regarding regional
development in Romania, there have been created 8 Development Regions,
corresponding to the NUTS II statistical level. These Regions, established through
voluntary co-operation of the counties (judets) do not enjoy legal status and are not
territorial-administrative units. More than half of Romanias towns (152 from 263)
have less than 20000 inhabitants and only 23 towns have a population exceeding
100000 inhabitants. Bucharest has more than 2 millions inhabitants. Urban population
represents 54.8% of total population.
Table 1: General Information on Romania and the Development Regions, 2000
Area, km2
Population
(thou)
238,391
22,456
Bacu, Botoani, Iai, Neam, Suceava,
Vaslui,
36,850
3,810
2. South-East
Brila, Buzu, Constana, Galai, Tulcea,
Vrancea
35,762
2,940
3. South
Arge, Clrai, Dmbovia, Giurgiu,
Ialomia, Prahova, Teleorman
34,453
3,480
4. South-West
Dolj, Gorj, Mehedini, Olt, Vlcea
29,212
2,410
5. West
Arad, Cara-Severin, Hunedoara, Timi
32,034
2,040
6. North-West
Bihor, Bistria-Nsud, Cluj, Maramure,
Satu-Mare, Slaj
34,159
2,850
7. Centre
Alba, Braov, Covasna, Harghita, Mure,
Sibiu
34,100
2,645
1,821
2,281
Region
Judets (counties)
ROMANIA
42 (Including Bucharest)
1. North-East
8. Bucuresti-Ilfov Bucureti, Ilfov
Source: National Commission for Statistics, 1999, and Institute for Economic Forecast, 2000.
2.2. Romanian Regional Disparities
Since the beginning of the transition period in 1990, Romanias economic situation has
declined significantly, and 1999 was the third consecutive year of real GDP decline,
4
which fell by 3.2% compared to 1998. This evolution led to a sharpening of the
disparities against EU-15.
DEVELOPMENT REGIONS
1 - NORTH-EAST
2 - SOUTH-EAST
3 - SOUTH-MUNTENIA
4 - SOUTH-WEST OLTENIA
5 - WEST
6 - NORTH-WEST
7 - CENTER
8 - BUCURESTI-ILFOV
PIATRA NEAMT
CLUJ NAPOCA
ALBA IULIA - Regional Development
ALBA IULIA
Agencies Headquarters
TIMISOARA
BRAILA
BUCURESTI
CALARASI
CRAIOVA
Table 2: Romania - Main economic indicators
Indicators
(percent)
1996
1997
1998
1999
2000
3.9
-6.1
-5.4
-3.2
0.91)
- annual average
38.8
154.8
59.1
45.8
- December-on-December
56.9
151.4
40.6
54.9
44.02)
6.6
8.9
10.3
11.5
11.83)
Real GDP growth rate
Inflation rate
Unemployment rate
Source: National Commission for Statistics, 1998, Statistical Year Book, and National
Commission for Statistics, 2000, Monthly Statistical Bulletin, 1-4.
Notes: 1) 2000, First Quarterly; 2) May, 2000 on May 1999; 3) At the end of May 2000.
The Romanian average GDP per capita at purchase power parity was 27% of the EU-15
average. There are significant disparities between the Development Regions. The most
5
developed one - the Bucharest-Ilfov Region - enjoyed a 38% of the EU-15 GDP per
capita average, while the poorest Region, the North-East, had an estimated 21% of this
average.
Table 3: GDP/capita in Romanian Development Region, 1999
Region
Romanian average GDP
per capita = 100%
EU-15 average GDP per
capita = 100%
ROMANIA
100
27
1. North-East
76
21
2. South-East
104
28
3. South
93
25
4. South-West
98
26
5. West
115
31
6. North-West
90
24
7. Centre
103
28
8. Bucuresti-Ilfov
142
38
Source: Calculations based on National Commission for Statistics, 1998, Statistical
Year Book, National Commission for Statistics, 2000, Monthly Statistical Bulletin,
2/feb., and Eurostat, 1999, l'Office Statistique des Communauts europennes
Luxembourg: L'largissement de l'UE Donnes cls sur les pays candidats, Memo
10/99, 7 dcembre.
In Romania, regional disparities have historical, geographical, cultural and economic
roots. These disparities, especially those economic, have expanded during transition
because, on the one hand, of substantial economic fall (at the end of 1999 GDP reached
only 75% of its 1989 level), and on the other hand of the firms' behaviour in an
economic environment with very high and long term inflation. In the same economic
environment, resources will be orientated to regions that offer the opportunity of a
rapid profit growth, and a rapid investment recapture (see, for details, Jula & Jula,
1998).
Moreover, the transition reveals the economic weakness of poor developed areas: the
strong dependence on a single industry, poor town planning and low localities
attractiveness, insufficient utilities infrastructure development a.s.o. The regions with
dominant rural areas are the poorest. They are strong dependent on agriculture and lack
6
a young and adult population (as in past decades they migrated to urban areas). In
accordance with Myrdal regional development theory (Myrdal, 1983), a development
of some centre (poles) produces a dual impact on surrounding areas. On the one hand,
there is a positive spread effect stimulation of development due to the demand of raw
materials, to the new technologies, innovation diffusion, etc. On the other hand, a
backwash effects the withdrawal from underdeveloped region of skilled labour,
capital and goods: the centre lure the qualitative elements of development, by filtering
the inflows and expel towards purlieus the cumbersome crumbs of their growth.
Unfortunately, the Romanian experience concerning the development role that urban
areas play within the locality network has recorded only the backwash effect: most of
Romanian cities have exhausted rather than generated energy and development in the
surrounding rural areas. This has deepened the urban-rural gap.
Significant disparities exist, however, within each Development Region. For example
in the Centre Development Region, Brasov and Sibiu counties are significantly more
urbanised and wealthier than the other four counties in the Region.
3. THE FACTORS OF REGIONAL COMPETITIVENESS
A recent study done for the European Commission (Pinelli, 1998) aimed at reducing
the competitiveness issue to its core elements through building a simple model
concerning the relationship between the regional GDP per capita and the main elements
that contribute to the specific indicators evolution. The approach followed was, firstly
to identify the main elements listed in the economic literature that are supposed to
explain variations in GDP per head between regions; secondly the making of a simple,
but statistically robust and observable indicator to represent it and, thirdly, to correlate
variations between these indicators across regions with variations in GDP per head as
well as GDP per person employed.
The conclusion of the study was that four factors emerged as being closely linked with
regional differences in the GDP measures:
- the structure of economic activity, which was simply represented as the distribution
of employment between agriculture, manufacturing, construction, market services
7
and non-market services; the regions with the highest levels of GDP per head tending
to have a relatively high concentration of employment in market services and/or
manufacturing;
- the extent of innovative activity, which was measured by the number of patent
applications; the best performing regions tending to be the source of more
applications than others;
- regional accessibility, which was measured by a new index of peripherality, which
implicitly includes the effects of variations in transport infrastructure; the regions
where GDP per head is above average tending to have better accessibility;
- the skills of the work force, which were measured by the relative numbers of people
aged 25 to 59 with high (university level or equivalent), medium (upper secondary
level qualifications) and low (basic schooling only) levels of education; the best
performing regions tending to have an above average proportion of relatively highly
qualified workers.
Econometrically, it is proved, using a simple linear regression equation, that these four
indicators explain almost two-thirds of the variation in GDP per capita between regions
in the European Union. It is considered that this result, however, needs to be interpreted
with a good deal of caution, because the correlation is only an average one and there
are many regions which diverge from the average (the model has not a reasonably high
value for R-squared).
For Romania, the correlation between the regional GDP per capita and the above
estimates of mentioned factors in the study of the European Commission is difficult to
make, mostly due to the difficulties related to the Romanian statistical system. Under
these circumstances, maintaining the above mentioned study approach, the
measurement of the innovation activity and of the accessibility degree is made using
other indicators. Thus, lacking the data concerning the regional patents demand, they
used data related to the research activity. The used indicator is the employed population
in research and development activities, compared to the whole amount of employed
population. Concerning the accessibility degree it is difficult to make a global indicator
to reflect the heavy infrastructure and the market size as it is recommended in the
European Commission study. Under these circumstances the accessibility is estimated
starting from the transportation infrastructure.
Education
level
Employment in
Table 4: Correlation coefficients between GDP per capita (regional values) and
competitiveness elements
Competitiveness elements
Correlation
coefficients
Agriculture and sylviculture
-0.735
Industry and construction
0.680
Services
0.492
Higher education
0.755
Gymnasium and secondary education
0.725
Primary education
-0.825
Employment in research-development activities
0.478
Railway transport infrastructure
0.391
Road transport infrastructure
0.089
The data presented in the Table 4 show that the Romanian competitiveness elements
keep the main trends recorded for the European Union. Under the caution of the
indicators significance used in calculations, we can assert that:
The regional GDP per capita is positively and strongly correlated with the degree of
employed population in manufacturing and construction (Pearson linear correlation
coefficient is +0.54). This value is, however, strongly influenced by the special
situation of the region that includes the country capital. Region VIII ((Bucharest +
Ilfov) is the most developed area in the country and has the best development factors. If
we let aside this region from our calculations, then the Pearson linear correlation
coefficient between GDP per capita and the degree of employment in the
manufacturing and construction is by +0.28, that indicates a regional distribution of
industry without too many disparities in the rest of the country.
9
The regional GDP per capita is positively and strongly correlated with the degree of
population employment in services (+0.93). This value is influenced only in a small
degree by the special situation of the region that includes the capital.
The regions with a population employment in agriculture are less economically
developed: the GDP per capita is negatively and strongly correlated with the degree of
population employment in agricultural activities (-0.93). Like the degree of
employment in services, this figure is influenced only in a small degree by the special
situation of the region that includes the capital.
On one hand, the regions that have a GDP per capita over the national average are a
focus for a highly trained work force (+0.92), and on the other hand, the correlation
between the regional development level and the extend of the elementary skilled
persons in the aged population is strongly negative (-0.96). The dimension of the
sample does not significantly influence this conclusion. Further to this context, there
are not significant disparities in the regional distribution of population with elementary
and college diplomas (the estimate ratio between standard deviation and series average
does not rise over 8-9%). But there are significant disparities in the regional
distribution of highly trained population.
In Romania there is a positive correlation between the regional distribution of the
innovation capacity and the GDP per capita (+0.82). The level of this indicator is
strongly influenced by the relatively strong level of the capital development (compared
to the rest of the territory). If we let aside from calculation the region that includes the
capital, then the correlation coefficient between the regional distribution of the
innovation capacity and GDP per capita distribution is only by +0.27. This proves that
the capital is a strong focus for the national innovation capacity.
Concerning the impact of accessibility degree over regional development we notice
both the positive relatively strong correlation between the railway network distribution
and the GDP per capita distribution and the positive connection between the
development level and the existing transportation territorial infrastructure elements.
10
Likewise, the reciprocal connections between the analysed factors match as a meaning
and to a great extend as intensity with the values calculated for European Union
regions.
Table 5: Correlation coefficients between regional competitiveness elements, 1997
1997
GDP EInd EAg EServ PHg PSec
GDP
EInd
EAg
1 0.54 -0.93
0.54
1 -0.71
-0.93 -0.71
0.93 0.92 0.96 -0.96 0.82 0.85 0.59
0.43 0.49 0.64 -0.57 0.53 0.58 0.44
1 -0.94 -0.95 -0.93 0.96 -0.91 -0.94 -0.71
EServ
0.93 0.43 -0.94
PHg
0.92 0.49 -0.95
0.99
PSec
0.96 0.64 -0.93
0.89 0.91
PPr
PPr ERs DRw DRo
1 0.99 0.89 -0.96 0.92 0.93 0.70
1 0.91 -0.98 0.96 0.97 0.75
1 -0.97 0.83 0.87 0.57
-0.96 -0.57 0.96 -0.96
ERs
0.82 0.53 -0.91
-0.97
1 -0.92 -0.95 -0.68
-0.98
0.92 0.96 0.83 -0.92
1 0.99 0.89
DRw
0.85 0.58 -0.94
0.93 0.97 0.87 -0.95 0.99
DRo
0.59 0.44 -0.71
0.70 0.75 0.57 -0.68 0.89 0.86
1 0.86
1
The symbols significations are the following:
Symbol
Signification
GDP
Gross Domestic Product per capita (1997) and by Development Regions,
calculated on the basis of the purchasing power parity
EAg
Employment in mining and quarrying, agriculture, sylviculture, forestry and
hunting, by Development Regions
EInd
Employment in manufacturing, electric and thermal energy, gas and water,
construction, by Development Regions
EServ
Employment in market services and non-market services, by Development
Regions
PHg
Higher education (university graduates) persons as share in population aged
20 59 years, by Development Regions
PSec
Gymnasium and secondary education persons as share in population aged 20
59 years, by Development Regions
PPr
Primary education persons as share in population aged 20 59 years, by
11
Symbol
Signification
Development Regions
ERs
Employment in research-development activities per 10,000 employed
persons, by Development Regions
DRw
Density of railway under operation (normal gauge lines) per 100 km2 of
territory, by Development Regions
DRo
Density of public roads per 100 km2 of territory, by Development Regions
For Romania, the strong correlation between the describing competitiveness elements
make impossible the building of a global econometric model, similar to the already
analysed one far the European Union regions. Econometrically speaking, this is due to
the strong connections between the describing variables that cause the multicollinearity
phenomenon. The effect of multicollinearity among explanatory variables is to increase
the standard errors of the regression coefficients and reduce the value of t-statistics,
thus making coefficients less significant (and possibly even insignificant). Furthermore,
the covariance between the regression coefficients of a pair of highly correlated
variables will be very high, thus making it difficult to interpret individual coefficients
(Ramanathan, 1992 and Pindyck and Rubinfeld, 1998): so, if two explanatory variables
are strongly correlated, their variance takes place simultaneously and it is difficult to
separate the individual effect on endogenous variable.
In spite of all these, the analysed data for Romania, both at county level and at
Development Region level demonstrates that the model built for the developed
economies from the European Union can be tailored for the regional analysis in the
transition economies, as well. In other words, even in weak structured economies, as
the Romanian economy, the GDP per capita is positively, relatively strongly correlated
with the degree of employment in manufacturing, in construction and in services, and
the region with a population mostly agrarian are less developed from an economic point
of view.
Likewise, the regions that have a GDP per capita over the national average are a focus
for the highly skilled work force while the correlation between the development level
and the share of elementary trained employable persons is strongly negative. One can
12
also state that in Romania there is a positive correlation between the regional
distribution of innovation capacity and the GDP per capita distribution. Concerning the
accessibility degree impact on regional development, we also notice that there is a
positive correlation between the road and railway infrastructure distribution and the
regional GDP per capita distribution.
At the same time, the reciprocal connections between the development elements are in
accordance as a meaning and to a great extent as intensity, to the values calculated for
European Union regions. Namely, in Romania, the regions where the population share
employed in the manufacturing, construction and services is high, these regions attract
a highly skilled work force and focus for the research activities. At the same time, the
mostly agrarian regions are strongly and positively correlated with an elementary
skilled population and have less heavy infrastructure elements.
References
Black J., 1997, Oxford Dictionary of Economics, Oxford University Press
Commission europenne, Direction gnrale Politique rgionale et Cohsion, 1999,
Sixime rapport priodique sur les rgions: rsum des principaux rsultats, Inforegio,
Fiche dinformation 04.02.99 FR, Luxembourg
Constantin D.-L., 1999, Regional Competition in Romania: Determinants and Policies,
38-th European Regional Science Association Congress, Dublin
Eurostat, 1999, l'Office Statistique des Communauts europennes Luxembourg:
L'largissement de l'UE Donnes cls sur les pays candidats, Memo 10/99, 7
dcembre, Luxembourg
Institute of Management Development, 1996, World Competitiveness Yearbook,
Lausanne: IMD
Jula D., Ailenei D., Jula N., Grbovean A., 1999, Economia dezvoltrii, Editura
Didactic i Pedagogic, Bucureti
Jula D., Jula N., 1998, The dynamics of regional disequilibria in the transition
economies: The Romanian case, 38-th European Regional Science Association
Congress, Vienna, Austria
Krugman P., 1994, Competitiveness: A Dangerous Obsession, Foreign Affairs,
March/April (volume 73, number 2)
13
Myrdal G., 1957, Economic Theory and Underdeveloped Regions, Duckworth London
National Commission for Statistics, 1998, Statistical Year Book, Bucharest
National Commission for Statistics, 2000, Monthly Statistical Bulletin, 1-2, Bucharest
North D.C., 1990, Institutions, Institutional Change and Economic Performance,
Cambridge University Press
Olson M., 1996 Big bills left on the sidewalk: why some nations are rich and others
poor, in Journal of Economic Perspectives, Volume 10, no.2
Pindyck R.S., Rubinfeld D.L., 1998, Econometric Models and Economic Forecasts, 4th
edition, McGraw-Hill
Pinelli D., 1998, Regional Competitiveness Indicators, Commission europene
Porter M.E., 1990, The competitive advantage of nations, Basingstoke: MacMillan
Ramanathan, R., 1992, Introductory econometrics with applications, The Dryden Press
Rhodes M., 1995, The Regions and the new Europe: Patterns in Core and Periphery
Development, Manchester: Manchester University Press
14