XXII. MEZINÁRODNÍ KOLOKVIUM O REGIONÁLNÍCH
VĚDÁCH. SBORNÍK PŘÍSPĚVKŮ
22nd
INTERNATIONAL COLLOQUIUM ON REGIONAL
SCIENCES.CONFERENCE PROCEEDINGS
Place: Velké Bílovice (Czech Republic)
June 12-16, 2019
Publisher: Masarykova univerzita (Masaryk University Press), Brno
Edited by:
Viktorie KLÍMOVÁ
Vladimír ŽÍTEK
(Masarykova univerzita / Masaryk University, Czech Republic)
Vzor citace / Citation example:
AUTOR, A. Název článku. In Klímová, V., Žítek, V. (eds.) XXII. mezinárodní kolokvium o regionálních
vědách. Sborník příspěvků. Brno: Masarykova univerzita, 2019. s. 1–5. ISBN 978-80-210-9268-6.
DOI.
AUTHOR, A. Title of paper. In Klímová, V., Žítek, V. (eds.) 22nd
International Colloquium on Regional
Sciences. Conference Proceedings. Brno: Masaryk University Press, 2019. pp. 1–5. ISBN 978-80-
210-9268-6. DOI.
Publikace neprošla jazykovou úpravou. / Publication is not a subject of language check.
Za správnost obsahu a originalitu výzkumu zodpovídají autoři. / Authors are fully responsible for the content and
originality of the articles.
© 2019 Masarykova univerzita
ISBN 978-80-210-9268-6 (online : pdf)
 
 
 
XXII. mezinárodní kolokvium o regionálních vědách Sborník příspěvků Velké Bílovice 12.–14. 6. 2019
 
37
DOI: 10.5817/CZ.MUNI.P210-9268-2019-4
HOW DID THE TIME GO WITH THE EU REGIONAL
COMPETITIVENESS INDEX? ANALYSIS OF CONTENT AND
METHODOLOGICAL CHANGES ACROSS EDITIONS
Jak šel čas s indexem regionální konkurenceschopnosti EU?
Analýza obsahových a metodických proměn napříč edicemi
MICHAELA STANÍČKOVÁ
LUKÁŠ MELECKÝ
Katedra evropské integrace
Ekonomická fakulta
VŠB - Technická univerzita Ostrava
Department of European Integration
Faculty of Economics
VŠB - Technical University of Ostrava
 Sokolská třída 33, 702 00 Ostrava, Czech Republic
E-mail: michaela.stanickova@vsb.cz, lukas.melecky@vsb.cz
Annotation
Why measuring regional competitiveness is so important? Because if you cannot measure it, you cannot improve
it. A quantitative score of competitiveness will facilitate the European Union (EU) in identifying possible regional
weaknesses together with factors mainly driving these weaknesses. This, in turn, will assist regions in the catching
up the process. In doing so, the paper looks at the impact of a variety of competitiveness factors, taking into
account regional differences which will affect their relative importance. In this case, the Regional Competitiveness
Index (RCI) can be used to help regions assess which aspects of competitiveness are the strongest and which ones
need improvement. RCI provides the first synthetic picture of the EU competitiveness at the level of NUTS 2 regions
representing, at the same time, a well-balanced plurality of different fundamental aspects. It takes into account
both social and economic aspects, including the factors which describe the short and long-term potential of the
economy. It is very important to understand the extent to which regions compete with each other, where this
competition comes from, and what factors determine a territorial, economic attractiveness. Therefore, regional
development strategies could use RCI to identify possible regional development priorities. By systematic analysis
of scientific literature comprehensive analysis of the EU regional competitiveness is performed.
Key words
composite index, edition, EU Regional Competitiveness Index, methodology, NUTS 2 region
Anotace
Proč je měření regionální konkurenceschopnosti tak důležité? Pokud není možné něco změřit, není možné to
zlepšit. Kvantitativní skóre konkurenceschopnosti usnadní Evropské unii (EU) postup při identifikaci možných
regionálních slabin spolu s faktory, které tyto slabiny nejvíce ovlivňují. To však napomůže regionům v rámci
konvergenčního procesu. Příspěvek se zabývá dopadem různých faktorů konkurenceschopnosti s přihlédnutím k
regionálním rozdílům, které ovlivní jejich relativní význam. V tomto případě lze použít Regionální index
konkurenceschopnosti (RCI) k tomu, aby regionům pomohl posoudit, které aspekty konkurenceschopnosti jsou
nejsilnější, a které je třeba zlepšit. RCI představuje první syntetický obraz o konkurenceschopnosti EU na úrovni
NUTS 2 regionů, který zároveň představuje vyváženou pluralitu různých základních aspektů. Zohledňuje sociální
i ekonomické aspekty, včetně faktorů, které popisují krátkodobý a dlouhodobý potenciál ekonomiky. Je velmi
důležité pochopit, do jaké míry si regiony navzájem konkurují, odkud tato konkurence pochází a jaké faktory určují
územní ekonomickou atraktivitu. Strategie regionálního rozvoje by proto mohly využít RCI k identifikaci možných
priorit regionálního rozvoje. Prostřednictvím systematické analýzy odborné literatury je provedena komplexní
analýza regionální konkurenceschopnosti v kontextu EU.
Klíčová slova
kompozitní index, edice, Index regionální konkurenceschopnosti EU, metodologie, NUTS 2 region
JEL classification: B41, O18, P51, R11, R58
XXII. mezinárodní kolokvium o regionálních vědách Sborník příspěvků Velké Bílovice 12.–14. 6. 2019
 
38
1. Introduction
A casual look at the geographical map of any territory reveals an uneven distribution of population across space.
These maps often feature small ‘nodal’ points of high population density, as opposed to the less inhabited
surrounding. Indeed, the uneven distribution of population and economic activity across space can be somewhat
attributed to natural causes. Certain climatic conditions and terrain configuration render chunks of territory
uninhabitable. On the other hand, fertile land along water sources provides favourable conditions for growth and
development of settlements. However, much of the pattern of uneven spatial distribution of economic activity
cannot be attributed solely to geographical reasons, but to a range of endogenous factors (De Bruyne, 2006). In
this context, it is the socio-institutional framework which becomes the determinant of the attractiveness of territory
for inflows of economic activity. As stated Danon (2014), given that the activity is unevenly distributed in function
of some endogenous factors, across a set of different territories, these inter-regional differences can induce
migrations of footloose factors of production between locations. In this context, as long as we assume a finite set
of factors of production with a limited level of mobility; territories can be considered to be competing for attraction
and retention of economic activity. Following this logic, then, a degree of successfulness in this competition can
be called territorial competitiveness (Camagni, 2002). Thus a competitive region, experiencing the prevalence of
agglomerative over dispersive forces, enjoys constant net positive migration of mobile factors of production. On
the other end of the scale, an uncompetitive region faces constant danger of desertification of footloose factors the
regional focus reflects the increasing consensus that they are the primary spatial units where increasing returns to
scale are created, while knowledge is generated and circulated, all resulting in creation of agglomerations of
economic activity (Huggins, 2003). In this context, it is affirmed the position that it is areas – sub-national
territories – rather than nation-states, which compete. A region is a useful choice principally because it is more or
less homogeneous, with a relatively similar institutional framework, economic and social structure within its
territory, while it usually doesn’t dispose of nation-state prerogatives. Therefore, the region has only recently
become a terrain of strenuous economic research. With the New Economic Geography integrating into the
mainstream, many spatial subjects, including territorial, or regional, competitiveness are being increasingly
inquired. In line with Krugman (2003), it is plausible to discuss competitiveness on a regional level, as a capacity
of territories to attract and retain mobile factors of production, which is an increasingly important subject in an
ever integrating global economy. However, this branch of economic geography is relatively underdeveloped, while
it even lacks a universally accepted definition and metrics. In the European Union (EU), the debate about
measuring and evaluation competitiveness at regional level resulted in the EU Regional Competitiveness Index
(RCI), a joint project between Directorate General (DG) Joint Research Centre and DG Regional Policy.
Generally, the framework of analysis presents the following sequence of steps, from the development of a
theoretical framework to the presentation and dissemination of the RCI. Each step is significant, but coherence in
the whole process is equally vital. Therefore, the special emphasis is put on theoretical background,
methodological application and finally to empirical results via systematic literature review approach investigating
research works on the issues of competitiveness. The paper highlights competitiveness and analyse its components
in the context of regional development. The study discusses background of regional competitiveness, address
related questions on the meaning and explanation of regional competitiveness and provides the regional
competitiveness factors. In this sense, the paper aims at aspects of regional competitiveness and methodological
restrictions and obstacles for measuring in the EU area what have an impact on the empirical analysis and practical
using for policy-makers and decision-makers. The purpose is thus measuring of competitiveness of the EU regions
at the NUTS 2 level by developing a composite index approach, and to look at methodological and empirical
changes over time 2010, 2013 and 2016.
2. Development of methodology across the RCI editions
The concept of competitiveness has in the last decades extended from the micro-level of firms to the macro-level
of countries. Between the two levels stands the concept of regional competitiveness which is the focus of the EU
RCI. The EU RCI is inspired by the Global Competitiveness Index (GCI) of the World Economic Forum (WEF).
Therefore, GCI has been the leading reference framework for RCI construction. This choice has been driven by
the fact that GCI is the most internationally recognised and acclaimed index in the field of competitiveness and its
framework covers a very comprehensive set of aspects relevant to competitiveness. There are, however, some key
differences that distinguish RCI from GCI due to RCI European and regional dimension. RCI has been published
every three years since 2010, coinciding with the European Commission reports on economic, social and territorial
cohesion. It covers regions at NUTS 2 level. RCI is based on indicators grouped in eleven pillars corresponding
to various aspects of regional competitiveness. Pillars are designed to capture short- as well as long-term
capabilities of the region. They are classified into three major groups: the pillars Institutions, Macro-economic
stability, Infrastructure, Health and Quality of Primary and Secondary Education are included in the first group
and represent the key basic drivers of all types of economies. As the regional economy develops, other factors
XXII. mezinárodní kolokvium o regionálních vědách Sborník příspěvků Velké Bílovice 12.–14. 6. 2019
 
39
enter into play for its advancement in competitiveness and are grouped in the second group of pillars – Higher
Education/Training and Lifelong Learning, Labour Market Efficiency and Market Size. At the most advanced
stage of development of a regional economy, key drivers for regional improvement are factors related to
Technological Readiness, Business Sophistication and Innovation, included in the third group. These aspects have
been selected in line with the definition of competitiveness used by the EC for RCI (the ability of a region to offer
an attractive and sustainable environment for firms and residents to live and work). According to authors of RCI
editions (Annoni, Kozovska, 2010; Annoni, Dijskstra, 2013; Annoni, Dijsktra, Gargano, 2017), this allows the
extension of traditional analysis of competitiveness to integrate perspectives of both businesses and inhabitants,
taking into account both business success and personal well-being.
RCI is the first measure providing a European perspective on the competitiveness of all NUTS 2 regions in the
EU. RCI makes it easy for a region to compare itself to all other regions, to spot regions with a similar level of
competitiveness, and to identify regions it could learn from. Through its eleven pillars, it assesses not only
aggregate competitiveness but also the strength and weaknesses of a region, using the main dimensions relevant
for competitiveness. Therefore, RCI offers a comprehensive picture on the situation of regions, allowing for a
cross-regional comparison among the EU Member States. As mentioned, RCI adopts and builds on the
methodology developed by the WEF in the form of GCI with some key differences, mainly due to RCIs European
and regional dimension. The guidelines which drove RCI analysis are major references of applied statistical
analysis (Knoke et al., 2002; Morrison, 2005; and OECD (2008). RCI is a composite indicator. As with all
combined measures, it has both advantages and disadvantages. It is a metric which quantifies with a single number
what is otherwise not directly measurable: the level of competitiveness of a region. The underlying phenomenon
remains intrinsically multidimensional. It is a combination of a large set of indicators which are observed,
statistically treated, weighted, and finally aggregated. RCI is, therefore, the result of a long list of subjective
choices. The heated debate within the scientific community regarding the pitfalls of summarising a multidimensional
concept in a single index – see for example Stieglitz’s Commission Report (Stiglitz et al., 2009) – is
not addressed here. RCI aims to provide a measure of the regional level of competitiveness by following two
interconnected principles: simplicity and transparency. Simplicity is driven by the general necessity of the
composite to be easily understood by a non-technical audience – policymakers, stakeholders, and citizens. But
simplicity did not prevail over technical soundness: appropriate statistical analyses drove RCI construction.
2.1 Framework of the dataset and geographical coverage
The set of indicators which populate each pillar is carefully chosen according to the literature review, experts’
opinion and data availability. Staníčková (2018) provides detailed information on indicators relevant for all RCI
editions. The dataset consists of total of 113 candidate indicators (90 construct indicator, and 23 candidate
indicators discarded) used RCI edition 2010, 2013 and 2016. Their code, description and type can be found in
original RCI publications, for more information see Annoni, Kozovska (2010), Annoni, Dijskstra (2013), Annoni,
Dijkstra, Gargano (2017). Data sources for RCI indicators are Eurostat, national statistical institutes, the World
Bank Worldwide Governance Indicators, Global Competitiveness Index (World Economic Forum), the Quality of
Government Index (Quality of Government Institute), OECD Programme for International Student Assessment
(PISA), Regional Innovation Scoreboard, Eurobarometer, Scopus indicators on publications by Science-Metrix,
and other indicators developed by DG Regio. Data for two RCI pillars are available at the country level only
(Macroeconomic stability; Quality of Primary and Basic Education), while the Institutions and Technological
Readiness pillar also include a national component.
One important note to the used working definition of competitiveness concept and tool in the form of RCI for its
measurement having a background in the database of indicators. It refers to a point in time and not to changes over
time. Likewise, all indicators included in RCI refer to the situation at one point in time and not to changes over
time. In line with a database of indicators must also be geographical coverage of analysis. The literature raises two
issues related to the selection of the appropriate regional level. First, competitiveness should be calculated for
functional economic regions. The second is that the region should have an important political and administrative
role. In most countries, however, functional regions are not administrative and vice-versa. Thus in practice, these
two recommendations can be rarely combined. RCI focuses on NUTS (Nomenclature of territorial units for
statistics) classification in the EU, specifically NUTS 2 regions are administrative or statistical regions which do
not take into account functional economic links. Even though RCI is not entirely comparable over time, due to
recurrent and often unavoidable revisions of regional indicators and NUTS classification, three RCI editions
provide a unique policy tool for monitoring and assessing regional competitiveness in the EU. For subsequent
relevant analysis of competitiveness across editions, a number of NUTS 2 regions must be the same. Due to the
fact, that number of evaluated NUTS 2 regions differs from RCI editions 2010, 2013 and 2016 – number of
evaluated NUTS 2 regions in RCI was unified based on the same classification in the fields of their names, codes
and geographical coverage (some regions were merged and joined across different RCI editions). The investigated
XXII. mezinárodní kolokvium o regionálních vědách Sborník příspěvků Velké Bílovice 12.–14. 6. 2019
 
40
territory of analysis is based on the EU NUTS classification, i.e. a hierarchical system for dividing up the EU
economic territory for several purposes, i.e. collection, development and harmonisation of the EU regional
statistics, socio-economic analyses of the regions and the framing of the EU regional policies (Eurostat, 2018).
Geographical coverage for RCI construction consists of 269 total regions (the official number of the EU28 NUTS
2 regions being 276 by Eurostat NUTS 2 2013 classification valid from January 1, 2015).
Related to RCI, it is necessary to note that edition 2010, 2013 and 2016 incorporates improvements and slight
modifications. However, these do not affect the overall structure of RCI. The reasons for the changes are multiple:
new indicators become available at the regional level, while others are not updated or no longer fit the statistical
framework of the index. Also, methodological improvements, especially between the first and the second RCI
editions, and changes in NUTS regions make these comparisons complex. Nevertheless, the method has not
changed substantially, and there is a high degree of continuity in the list of indicators. Changes in RCI editions are
presented in Table 1.
Tab. 1: Comparison of RCI editions
Edition Data Reference Year Geographical Level/Pillars Indicators
RCI
2010
Most recent data for all
indicators, temporal range
differs: individual years 2000,
2005, 2006, 2007, 2008, 2009
and their variable combinations
in averages
National: country level (Institutions, Macroeconomic
Stability, Quality of primary & secondary education,
Technological readiness-part enterprises, not
households)
Regional: NUTS 2 level: all the other pillars; 268
NUTS 2 regions – Eurostat classification 2010
Indicators:
81
candidate,
69 construct
RCI
2013
Most recent data for all
indicators, temporal range
differs: individual years 2006,
2009, 2010, 2011 and their
variable combinations in
averages
National: country level (Institutions, Macroeconomic
Stability, Quality of primary & secondary education,
Technological readiness-part enterprises, not
households)
Regional: NUTS 2 level: all the other pillars; 273
NUTS 2 regions – Eurostat classification 2010
Indicators:
80
candidate,
73 construct
RCI
2016
Most recent data for all
indicators, temporal range
differs: individual years 2009,
2011, 2013, 2014, 2016 and
their variable combinations in
averages
National: country level (Institutions, Macroeconomic
Stability, Quality of primary & secondary education,
Technological readiness-part enterprises, not
households)
Regional: NUTS 2 level: all the other pillars; 275
NUTS 2 regions – Eurostat classification 2013
Indicators:
79
candidate,
74 construct
Source: Annoni, Kozovska (2010); Annoni, Dijskstra (2013); Annoni, Dijkstra, Gargano (2017); analysed and elaborated by
Staníčková (2018)
2.2 Statistical assessment of RCI
Constructing a composite is a multidisciplinary exercise which involves expertise both in the concept to be
measured and in the statistical techniques used to assess the developing methodology. Statistical assessment of
RCI is carried out with a twofold intention, i.e. to assess the quality of each indicator selected to populate RCI
framework: it includes the treatment of missing values and outliers (univariate analysis, i.e. indicator by indicator),
and to verify whether the set of indicators within each dimension is jointly consistent (multivariate analysis, i.e.
on each pillar as a whole). The key driver for RCI computation has been to keep it simple, to be easily understood
by non-statisticians, and at the same time consistent (Annoni, Kozovska, 2010).
The former allows for detecting possible problems with: i) missing data; ii) distribution asymmetry and outliers
and iii) different measurement scales. These problems are addressed by adopting: i) specific imputation methods;
ii) power-type transformations to correct for skewness; iii) standardisation. The multivariate analysis is carried out
at the pillar level on the set of indicators as a whole. The aim is to assess their contribution in describing the latent
dimension behind each pillar. Anomalous indicators are in some cases detected and excluded from further analysis.
A number of the candidate and construct indicators differs across RCI editions 2010, 2013 and 2016. Following
the structure of statistical assessment presented in RCI, a separate discussion of each of the eleven pillars was
outlined in each edition. For each pillar, the chosen indicators are individually analysed by univariate statistical
methods and as a whole by the multivariate approach. The indicators used have a direct positive relationship with
competitiveness, i.e. the higher their value, the higher the level of competitiveness. Whenever necessary, original
indicators have been reversed. Multivariate analysis has been used to verify the existence of a single latent
dimension. In few cases indicators which do not describe this universal dimension, underlying the specific pillar,
have been discarded. The geographical distribution of the pillar sub-score, computed as a simple average of the
transformed/standardised indicators, is shown. Sub-scores are presented as min-max normalised scores (as a
XXII. mezinárodní kolokvium o regionálních vědách Sborník příspěvků Velké Bílovice 12.–14. 6. 2019
 
41
percentage) and are divided into six classes, with high values associated with high competitiveness. Tables with
corresponding sub-scores and the regions’ ranks are also presented in each edition in a pillar by pillar analysis.
Construction of final RCI consists of a step-by-step aggregation process. The first step consists of computing the
scores for each RCI dimension as a simple arithmetic average of the transformed and normalised indicators which
passed the Principal Component Analysis (PCA) test. It is important to remark that RCI is not based on PCA
weights. The derivation of weights through principal component or other latent variable models is neither
straightforward nor transparent. Pure statistical approaches may lead to inappropriate normative results such as,
for example, the assignment of negative weights to some dimensions (Decancq, Lugo, 2009). High correlation
within each pillar also reduces the possibility of compensability across indicators, where compensability is
understood as the undesirable offsetting of low scores in some indicators with high scores in others.
The second step consists of computing the scores for the three groups of dimensions – Basic, Efficiency and
Innovation – as arithmetic means of the dimension scores. For each pillar, RCI sub-scores have computed as a
simple average of the standardised (z-scores) indicators values. Scores at the pillar group level (sub-indices) are
computed as an average of the corresponding sub-scores. A sub-index is thus calculated for each of the three pillar
groups (Basic, Efficiency and Innovation) as the simple average of the pillar scores belonging to each of the pillar
groups.
For each region i the sub-scores associated to the dimension groups are:
5
basic
1
1
RCI score
5 j
(i) (i, j)

  (1)
where score (i,j) is the score assigned to region i for dimension j; basicRCI presents sub-index of RCI for i-th
region; i is EU28 NUTS 2 region; i {1 = AT11, … , 269 = UKN0}; j is pillar of competitiveness relevant for
dimension of sub-index basicRCI ; j {1 = Institutions, 2 = Macroeconomic stability, 3 = Infrastructure, 4 =
Health, 5 = Quality of primary and secondary education}.
8
efficiency
6
1
RCI score
3 j
(i) (i, j)

  (2)
where score (i,j) is the score assigned to region i for dimension j; efficiencyRCI presents sub-index of RCI for i-th
region; i is EU28 NUTS 2 region; i {1 = AT11, … , 269 = UKN0}; j is a pillar of competitiveness relevant for
the dimension of sub-index efficiencyRCI ; j {6 Higher education and training and Lifelong Learning, 7 Labour
market efficiency, 8 Market size}.
11
innovation
9
1
RCI score
3 j
(i) (i, j)

  (3)
where score (i,j) is the score assigned to region i for dimension j; innovationRCI presents sub-index of RCI for i-th
region; i is EU28 NUTS 2 region; i {1 = AT11, … , 269 = UKN0}; j is pillar of competitiveness relevant for
dimension of sub-index innovationRCI ; j {9 Technological readiness, 10 Business sophistication, 11 Innovation}.
In the last step, the RCI score is computed as a weighted average of three sub-scores:
basic basic efficiency efficiency innovation innovationRCI RCI RCI RCI(i) w (i)+ w (i)+ w (i) (4)
basic efficiency innovation 1w + w + w  (5)
0 1w ,  (6)
where ( )iRCI presents composite weighted index RCI for i-th region; w is normalised weight (based on the stage
of development) of i-th region for j-th pillar of competitiveness representing relevant sub-index.
Final RCI score is the result of a weighted aggregation of three sub-indices, and the set of weights is chosen
according to the development stage of the region. For final aggregation, RCI follows the approach that the WEF
adopts for GCI with the aim of taking into account the level of heterogeneity of the EU regions, especially after
the enlargements. GCI takes into account the stage of development (SoD) of a country and accordingly assigns a
different weighting scheme to groups of pillars (Schwab, Porter, 2007). Given that some variability across the
development stages of the EU NUTS 2 regions is expected, a similar approach is adopted for RCI. The set of
weights adopted for aggregating the sub-indices depend on the level of development of the regions, classified into
the medium, intermediate and high stage on the basis of their GDP value. Table 2 and Table 3 show weights
assigned to SoD based on GDP threshold. On the basis of GDP thresholds, the EU NUTS 2 regions were classified
into different SoD. It is evident that in the RCI 2010 edition, only three SoD categories were recognised. In RCI
XXII. mezinárodní kolokvium o regionálních vědách Sborník příspěvků Velké Bílovice 12.–14. 6. 2019
 
42
2013 and 2016 editions, the weighting system and the classification of the regions into SoD have been slightly
modified, also following the suggestions by the WEF team in charge of GCI (Annoni, Dijskstra, 2013). Five
classes, instead of three of RCI 2010 edition, are used to allow for a smoother change in the weighting values
across SoD. RCI does not have any transition stages which are instead used in WEF-GCI with the country-specific
set of weights. By adding two more classes, RCI tries to cope with this issue.
Medium SoD is associated with regional economies primarily driven by factors such as lower skilled labour and
basic infrastructures. Aspects related to good governance and quality of public health are considered basic inputs
in this framework. Intermediate SoD is characterised by labour market efficiency, quality of higher education and
market size, factors which contribute to more sophisticated regional economies and more significant potential for
competitiveness. In the high SoD, factors related to innovation, business sophistication and technological readiness
are necessary inputs for innovation-driven regional economies. The level of competitiveness of more developed
economies, on the other hand, takes into account to a more significant extent their innovation capability as a critical
driver for their advancement (Annoni, Kozovska, 2010). The weighting scheme of pillar groups has the effect of
not penalising regions on factors where they lay too far behind. RCI message is then more constructive: it provides
a measure of competitiveness which allows for fair comparison of the EU regions and highlights realistic areas of
improvement. As for almost every CI, the procedure followed for the setting up of RCI is affected by a certain
degree of subjectivity.
Tab. 2: RCI 2010 normalised weights
Average GDP/p.c. in PPS
expressed as an index (EU =100)
SoD SoD description
Weight-
Basic
Weight-
Efficiency
Weight-
Innovation
Sum of
Weights
<75 1 Medium (M) 0.4000 0.5000 0.1000 1.0000
≥ 75 and < 100 2 Intermediate (I) 0.3000 0.5000 0.2000 1.0000
≥ 100 3 High (H) 0.2000 0.5000 0.3000 1.0000
Source: Annoni, Kozovska (2010); elaborated by Staníčková (2018)
Tab. 3: RCI 2013–2016 normalised weights
Average GDP/p.c. in
PPS expressed as an
index (EU =100)
SoD SoD description
Weight-
Basic
Weight-
Efficiency
Weight-
Innovation
Sum of
Weights
<50 1 Medium (M) 0.3500 0.5000 0.1500 1.0000
[50–75) 2 Medium /Intermediate (M/I) 0.3125 0.5000 0.1875 1.0000
[75–90) 3 Intermediate (I) 0.2750 0.5000 0.2250 1.0000
[90–110) 4 Intermediate/High (I/H) 0.2375 0.5000 0.2625 1.0000
≥ 110 5 High (H) 0.2000 0.5000 0.3000 1.0000
Source: Annoni, Dijskstra (2013); Annoni, Dijkstra, Gargano (2017); elaborated by Staníčková (2018)
3. Time comparison analysis of the RCI results: Regional evidence of the EU
competitiveness
It has been initially assumed that the suggested measure of competitiveness can take both positive and negative
scores. Its positive score indicates that a region is competitive, whereas a negative one informs about the absence
of competitiveness. So, have regions improved their competitiveness in time?
Results show the level of heterogeneity in RCI results in time, a certain level of homogeneity across the EU is
expected and the diversity suggests substantial differences in the sophistication of regional economies across and
within countries. RCI shows a more polycentric pattern with substantial capital and metropolitan regions in many
parts of Europe. Similarly competitive regions surround some capital regions, but in many countries, the regions
neighbouring the capital are less competitive. The substantial disparities within several countries also highlight the
need for regional analysis and the limits of a purely national approach. RCI scores have been modified to mimic
the regions used in 2010, 2013 and 2016 editions, based on used indicators and changes in NUTS classification
(especially numbers, names and codes on regions) and combine regions with different stages of development which
makes the comparison every more fraught and captures results more relevantly as possible – this makes it possible
to compare RCI and RRI results. Table 4 and Table 5 show scores and ranks of the best and worst EU28 NUTS 2
regions. In all RCI editions, most of the top regions host either capitals or large metropolitan areas which help to
boost the region’s competitiveness; to this group belong only regions of the so-called old EU Member States, i.e.
United Kingdom, Netherlands, Denmark, Sweden and Finland, and also France and Belgium too). At the other
end of the scale, we find Greek, Portuguese, Bulgarian and Romanian regions.
XXII. mezinárodní kolokvium o regionálních vědách Sborník příspěvků Velké Bílovice 12.–14. 6. 2019
 
43
Tab. 4: RCI TOP 10 position of NUTS 2 regions during reference periods
NUTS 2
RCI 2010
NUTS 2
RCI 2013
NUTS 2
RCI 2016
Score Rank Score Rank Score Rank
NL31 1.253 1 NL31 1.358 1 UKH2 1.214 1
FI1B 1.134 2 UKH2 1.192 2 UKH3 1.214 1
DK01 1.130 3 UKH3 1.192 2 UKI3 1.214 1
NL32 1.116 4 UKI3 1.192 2 UKI4 1.214 1
SE11 1.081 5 UKI4 1.192 2 UKI5 1.214 1
FI1C 1.031 6 UKI5 1.192 2 UKI6 1.214 1
NL33 1.024 7 UKI6 1.192 2 UKI7 1.214 1
FR10 1.017 8 UKI7 1.192 2 UKJ1 1.150 8
NL41 0.993 9 UKJ1 1.174 9 NL31 1.149 9
BE24 0.969 10 SE11 1.149 10 SE11 1.138 10
Source: calculated and elaborated by Staníčková (2018)
Tab. 5: RCI BOTTOM 10 position of NUTS 2 regions during reference periods
NUTS 2
RCI 2010
NUTS 2
RCI 2013
NUTS 2
RCI 2016
Score Rank Score Rank Score Rank
BG33 –1.294 260 EL61 –1.337 260 BG34 –1.334 260
RO12 –1.294 260 RO41 –1.360 261 RO31 –1.341 261
EL51 –1.311 262 RO12 –1.362 262 RO41 –1.345 262
RO41 –1.369 263 EL30 –1.371 263 EL62 –1.364 263
EL64 –1.376 264 EL64 –1.376 264 EL63 –1.410 264
RO22 –1.385 265 EL42 –1.403 265 EL65 –1.443 265
BG31 –1.387 266 BG34 –1.403 266 BG31 –1.445 266
EL52 –1.465 267 EL54 –1.417 267 EL64 –1.446 267
PT20 –1.485 268 RO22 –1.479 268 EL51 –1.490 268
EL63 –1.511 269 BG31 –1.481 269 RO22 –1.494 269
Source: calculated and elaborated by Staníčková (2018)
In relation to findings from RCI 2010 edition, the most competitive EU regions are the three Dutch provinces that
cover the Randstad (Utrecht, North and South Holland); Hovedstaden which includes Copenhagen; London;
Stockholm; South Finland which includes Helsinki; Ile de France which includes Paris; North Brabant; and the
UK region Berkshire, Buckinghamshire and Oxfordshire. In many countries, the capital region is far more
competitive than the other regions in the same country, and many countries have highly heterogeneous RCI scores.
Based on findings from RCI 2013 edition, the best group includes again Utrecht, the highest competitive region in
both editions, the London area and the area including Oxford, the two Netherland regions of Noord and Zuid
Holland which comprise Amsterdam, the Danish region Hovedstaden (including Copenhagen), Stockholm and Île
de France (including Paris). The new entries in the top-ten are the Frankfurt region (Darmstadt) and the Surrey,
East & West Sussex in the United Kingdom. It is striking that seven out of the top ten are either capital regions or
regions including large cities. At the other end of the competitiveness scale, we find some regions which are
unfortunately steadily worsted performers. These are the Bulgarian region Severozapaden, the Greek region Notio
Aigaio, and two southern Romanian regions Sud-Est and Sud-Vest Oltenia. Findings from RCI 2016 edition, show
a polycentric pattern, with capital regions and regions with large metropolitan areas scoring the highest on the
index. In most of north-western Europe, such regions tend to generate spillovers (i.e. boost the competitiveness of
the neighbouring regions), while this is less the case in eastern and southern states. The best performers in RCI
2016 edition were several UK regions (London and its commuting zone; Berkshire, Buckinghamshire and
Oxfordshire; Surrey, East and West Sussex; Hampshire and the Isle of Wight), Utrecht (the Netherlands),
Stockholm (Sweden), Hovedstaden (Denmark), Luxembourg, Ile de France (France) and Oberbayern (Germany).
The bottom performers were several Romanian regions (Sud-Muntenia, Sud-Vest Oltenia, Sud-Est), Greek regions
(Ionia Nisia; Dytiki Ellada; Peloponnisos; Sterea Ellada; Anatoliki Makedonia, Thraki), the Bulgarian region of
Severozapaden and French Guyane. As in the previous two editions, the geographical distribution of
competitiveness features wide variations not only between but also within countries. Compared with the last two
editions (2010 and 2013), Malta and several regions in France, Germany, Sweden and the UK have improved,
while scores went down in Cyprus and some regions in Greece, Ireland and the Netherlands. In eastern EU regions,
competitiveness has mostly remained stable. High within-country variations were often observed due to a
significantly outperforming capital region. This gap was unusually wide in Romania, Greece, Slovakia, Bulgaria
and France, and relatively small in the United Kingdom, Austria and Belgium. However, the capital region was
not the most competitive region in Germany, Italy or the Netherlands.
XXII. mezinárodní kolokvium o regionálních vědách Sborník příspěvků Velké Bílovice 12.–14. 6. 2019
 
44
4. Conclusion
Comparing RCI over time is complicated because each edition 2010, 2013 and 2016 incorporates improvements
and slight modifications. These do not affect the overall structure of RCI, but they limit the possibilities to measure
change over time. The reasons for the changes are multiple: new indicators become available at the regional level,
while others are not updated or no longer fit the statistical framework of the index. Also, methodological
improvements, especially between the first and the second RCI editions, and changes in NUTS regions make these
comparisons complex. Nevertheless, the method has not changed substantially, and there is a high degree of
continuity in the list of indicators. Necessary to note that, in the paper, scores for RCI have been modified to mimic
the regions used in 2010, 2013 and 2016 editions, based on used indicators and changes in NUTS classification
and combine regions with different stages of development which makes the comparison every more fraught and
captures results more relevantly as possible (with respect to negatives of RCI). Although this approach allows
comparison of RCI editions, it is not possible to identify whether the differences in the score is due to changes in
data or changes in method. In some cases, it may be primarily due to changes in values of indicators included; in
other cases, it may be mainly due to changes in methodology and indicator selection. Changes in a region’s ranking
over time are not always meaningful. Rankings are based solely on the sequence of the scores and do not take into
account the actual differences between scores. Comparison of RCI ranks is not feasible because RCI implements
some improvements and modifications that, even if not affecting the overall index structure, make the direct score
comparison not meaningful. Ranking comparability over time is also troublesome as ranks are mutually dependent.
A change in the ranking may be due to a minimal difference in scores, which is not significant. This is often the
case for scores around the average. Analysing the significant changes in the scores, as opposed to the rankings,
over time can be highly informative. RCI has three editions and includes updated and more data together with
method refinements. It is important to note that the increased number of indicators, but also their internal changes
and the methodological improvements reduce the comparability over time. However, despite all its shortcomings
in all editions, RCI proved to be a robust way to summarise many different indicators into one index.
Literature
[1] ANNONI P., DIJKSTRA, L., (2013). EU Regional Competitiveness Index 2013. Luxembourg: Publications
Office of the European Union. ISBN 978-92-79-32370-6.
[2] ANNONI P., DIJKSTRA, L., GARGANO, N., (2017). EU Regional Competitiveness Index 2016. Working
Paper WP 02/2017. European Commission: Directorate-General for Regional and Urban Policy.
[3] ANNONI, P., KOZOVSKA, K., (2010). EU Regional Competitiveness Index 2010. Luxembourg: Publications
Office of the European Union. ISBN 978-92-79-15693-9.
[4] CAMAGNI, R., (2002). On the concept of territorial competitiveness: Sound or misleading? Urban Studies,
vol. 39, no. 13, pp. 2395-2411. ISSN 0042-0980. DOI 10.1080/0042098022000027022.
[5] DANON, M., (2014). Constructing a Novel Competitiveness Index for European Regions. GREDEG Working
Papers 2014–42. Nice: University of Nice Sophia Antipolis.
[6] DE BRUYNE, K., (2006). The location of economic activity. First versus second nature core-periphery
theories. Review of Business and Economics, vol. 51, no. 1, pp. 75-104. ISSN 2311-0279.
[7] DECANCQ, K., LUGO, M., (2009). Setting weights in multidimensional indices of well-being and
deprivation. OPHI working paper no. 18. Oxford: University of Oxford.
[8] EUROSTAT., (2018). NUTS – Nomenclature of Territorial Units for Statistics. Overview [Online]. Available
at www: ˂http://ec.europa.eu/eurostat/web/nuts/ overview˃.
[9] HUGGINS, R. (2003). Creating a UK competitiveness index: Regional and local benchmarking. Regional
Studies, vol. 37, no. 1, pp. 89-96. ISSN 0034-3404. DOI 10.1080/0034340022000033420.
[10]KNOKE, D., BOHRNSTEDT, G.W., POTTER MEE, A., (2002). Statistics for Social Data Analysis.
Belmont: Wadsworth Publishing. ISBN 978-0875814483.
[11]KRUGMAN, P., (2003). Second winds for industrial regions. In Coyle D., Alexander W., Ashcroft, B., (eds).
New Wealth for Old Regions. Oxford: Princeton University Press, pp. 35-47. ISBN 9780691122564.
[12]MORRISON, D.F., (2005). Multivariate Statistical Methods. Hampshire: Duxbury Press. ISBN 978-
0534387785.
[13]ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT., (2008). Handbook on
Construction Composite Indicators. Methodology and User Guide. Paris: OECD. ISBN 978-92-64-04345-9.
[14]SCHWAB, K., PORTER, M. E., (2007). The Global Competitiveness Report 2007–2008. Geneva: WEF.
ISBN 978-92-95044-09-8.
XXII. mezinárodní kolokvium o regionálních vědách Sborník příspěvků Velké Bílovice 12.–14. 6. 2019
 
45
[15]STANÍČKOVÁ, M., (2018). EU Competitiveness and Resilience: Evidence-based on Regional Level. SAEI,
vol. 51. Ostrava: VSB-TU Ostrava. ISBN 978-80-248-4168-7.
[16]STIGLITZ, J., SEN, A., FITOUSSI, J.P., (2009). Report by the Commission on the Measurement of Economic
Performance and Social Progress. Paris: OECD.
The paper is supported by the grant No. 17-23411Y of the Czech Science Agency, the SGS project (SP2017/111)
of Faculty of Economics, VŠB-TUO and the Operational Programme Education for Competitiveness – Project
No. CZ.1.07/2.3.00/20.0296.