Professional Documents
Culture Documents
Banchmarking Urban Competitivness Europe PDF
Banchmarking Urban Competitivness Europe PDF
Banchmarking Urban Competitivness Europe PDF
Cities
journal homepage: www.elsevier.com/locate/cities
a r t i c l e i n f o a b s t r a c t
Article history: The current context of the global economy and a shortage of capital is putting cities all over the world under
Received 13 March 2015 great pressure and increasing competition between them. The competitive position of a city does not
Received in revised form 29 May 2015 remain stable over time: it changes due to both internal and external factors such as the appearance of
Accepted 1 June 2015
new technologies, new competitors and changes in the preferences and desires of its target groups. Cities
Available online 18 June 2015
therefore need to understand the strengths and weaknesses that inuence not just their own ability to com-
pete but also that of their competitors. This paper seeks to benchmark the competitiveness of European
Keywords:
cities as locations for businesses and in terms of their ability to attract investment. Urban competitiveness
Globalisation
Urban competitiveness
is a complex, multidimensional issue, so a scale of measurement has been created based on a synthetic
Innovation index called the Urban Competitiveness Index (UCI), which comprises various sub-indices representing
Benchmarking its various dimensions. In the present study we focus on the basic dimension, the efciency dimension
Index and the innovation dimension. The study covers 159 cities located in 26 European Union countries.
2015 Elsevier Ltd. All rights reserved.
1. Introduction territories and the formation of global cities (Sassen, 1991, 1994,
2007), in a new trend towards the formation of mega-regions/
The globalisation of the worlds economy at the end of the 20th mega-cities (Florida, 2009; Florida, Gulden, & Mellander, 2007)
century gave rise to far-reaching changes in cities all over the world. and an increase in cross-border ows that connect cities at
Those changes were not just economic but also social, cultural, polit- different levels of the urban hierarchy (Camagni, 2004), in which
ical and structural. This process was accompanied by a concurrent hierarchical forms and forms of cooperation are mixed in links
process of localisation, giving rise to a global/local duality; global between cities. Thus, cities face the challenge of redesigning
opportunities arise that must be developed using local capabilities themselves as systems that can handle difculties and adapt
and initiatives, thus increasing the economic and political power quickly and effectively to the new challenges and opportunities
of both cities and the regions to which they belong. Combined with entailed by a highly dynamic global environment. That redesign
technological advances and the advent of new information and com- must enable them to shift from industrial societies to information,
munication technologies (ICTs), this has turned the worlds econ- knowledge and/or learning societies.
omy into a global, information economy (Castells, 1997). It is These are societies in which individual and collective abilities to
global because production, consumption and the circulation of cap- obtain knowledge and learn enable processes of regeneration and
ital are all organised on a global scale either directly or via a network economic revitalisation to be undertaken with some assurance of
of links between the main economic actors. It is information-based success. Thus, globalisation must be factored into city manage-
because productivity does not depend solely on conventional pro- ment, and cities nd themselves competing with one another to
duction factors but also on the ability to generate and process infor- attract anything that can create wealth (Darchen & Tremblay,
mation and knowledge strategically (Dosi, Freeman, Nelson, 2010; Zenker, 2009) and employment.
Silverberg, & Soetel, 1988; Foray & Freeman, 1992). This increase in competition between cities means that their tar-
But globalisation has also produced changes in territorial struc- get groups pick whichever city best suits their wishes and needs,
tures: its direct effects can be seen in the appearance of a number that is, whichever offers the greatest added value. Cities therefore
of global systems that operate in a context of worldwide ows and need to identify their competitors and determine where their com-
communication. Indirectly, this has resulted in a restructuring of petitive advantage lies if they are to achieve growth and economic/
social protability. This means analysing, interpreting and assessing
Corresponding author. their positions relative to their competitors in regard to a number of
E-mail addresses: lucia.saez@ehu.es (L. Sez), inaki.perianez@ehu.es
characteristics or attributes which cities may or may not possess. In
(I. Periez). short, cities need to engage in benchmarking to help their managers
http://dx.doi.org/10.1016/j.cities.2015.06.002
0264-2751/ 2015 Elsevier Ltd. All rights reserved.
L. Sez, I. Periez / Cities 48 (2015) 7685 77
identify competitors, establish competition proles and determine The concept of urban competitiveness was dened by Lever and
where their competitive advantage may lie on the one hand, and Turok (1999: 792) as . . . the degree to which cities can produce goods
to develop distinctive positioning strategies on the other. Various and services which meet the test of wider regional, national and inter-
authors (Bronisz, Heijman, & Miszczuk, 2008; Cabrero, 2012; national markets, while simultaneously increasing real incomes,
Cabrero, Orihuela, & Ziccardi, 2005; Chesire, Carbonaro, & Hay, improving the quality of life for citizens and promoting development
1986; Freudenberg, 2003; Huggins, 2003; Jiang & Shen, 2010; in a manner which is sustainable. In this denition the authors take
Lukovics & Lengyel, 2006; Suan, 1993; Villaverde, 2007) and insti- into account not just the nancial protability which a city needs to
tutions (Beacon Hill, 2007; European Union Regional Policy, 2011; be competitive, but also social protability. This mean that cities
IMD, 2008; Ni, 2012; WEF, 2012) have drawn up studies for bench- like the regions in which they are located can either help rms to
marking cities from a number of perspectives (outcomes, inputs and be competitive or become the main obstacle to their being so.
a combination of the two) and using different indicators and meth- Other authors, such as Porter (1991, 1995, 1996), Moori-Koening
ods. However, none of these studies refers exclusively to cities in and Yoguel (1998), Begg (2002) and Sobrino (2002), hold that com-
Europe, where the expansions of the EU in 2004 and 2007 have petitiveness is a process of generating and disseminating compe-
resulted in a union of territories, countries, regions and cities with tencies which depends not only on microeconomic factors
widely varying levels of economic development. (businesses) but also on the ability of areas to offer features that
The main contribution of this study is a benchmarking of facilitate economic activities. In other words, the idea is to create
European cities in terms of how competitive they are when it a physical, technological, social, environmental and institutional
comes to getting businesses to locate in them and attracting setting conducive to attracting and developing economic activities
investment. Urban competitiveness is a complex, multidimen- that can create wealth and jobs. Cities can promote or create such
sional issue, so a scale of measurement has been drawn up conditions and thus turn competitiveness into a determinant factor
based on a synthetic index, referred to here as the Urban for their economic development. This may even lead to cities com-
Competitiveness Index (UCI). This index is made up of peting directly with one another through actions and policies
sub-indices representing its various dimensions, namely, the basic implemented by their authorities. In a globalised setting the most
dimension, the efciency dimension and the innovation dimen- economically successful cities are those that act with some degree
sion. To that end, 31 indicators are used, most of which are drawn of autonomy from their national economies and are able to compete
from Eurostat (Appendix A), in relation to these three dimensions. internationally. The performance of the economy of a country thus
Our sample comprises 159 cities from 26 European Union coun- depends on the actions of its cities on the international stage, but
tries, classed as LUZs (Large Urban Zones), which means that they those cities are less dependent on the performance of their domes-
have a population of at least 100,000. tic markets as a whole (Camagni, 2002; Lever, 1999).
The concepts of competition and competitiveness are linked but
are not synonymous: the latter is a consequence of the former. It is
2. The concept of competitiveness applied to cities the degree of competition that currently exists between cities at
domestic and international levels that forces them to be competi-
In recent years academic studies have paid more and more tive. Thus, competitiveness becomes a determining factor for the
attention to the concept of territorial (regional and urban) compet- economic development of cities; when a city is capable of creating
itiveness (Ache & Andersen, 2008; Begg, 1999; Buck, 2005; a setting that is conducive to and suitable for competitiveness it
Cabrero, 2012; Cabrero et al., 2005; Camagni, 2002; Gardiner, can be referred to as a competitive city. Accordingly, a city (and by
Martin, & Tyler, 2004; Gordon, 1999; Jensen-Butler, 1999; Jiang & extension a country or region) can be seen as competitive insofar
Shen, 2010; Lengyel, 2004; Lever, 1999; Lukovics & Lengyel, as its production activities and its public, social and private organi-
2006; Martin, 2003; Turok, 2004). Their interest has also extended sations as a whole are effective, efcient, enterprising and innova-
to regional and urban political discourses. Several international tive. This means that they need to have the support of
organisations (European Commission, 2011; IMD, 2008; OECD, infrastructure, amenities, human capital and the institutions
2005) have dened the term competitiveness in regard to terri- required to put their competitive advantage to good use and make
tories. They coincide at least in positing that competitiveness refers them truly competitive. Not all cities compete in the same condi-
on the one hand to the actual, physical setting and on the other tions, and not all have the same resources or capabilities to deal with
hand to the concept of productivity. Accordingly, getting a higher competitors.
yield from natural resources, labour and capital is essential, but There are cities which have unique attractions which are hard
not necessarily sufcient in itself, to make a territory more com- to imitate, and which distinguish themselves and furnish them
petitive. In political discourse, the idea that territories, regions with a competitive edge. The rest must be able to develop their
and cities compete with one another and that there is room for resources and generate their own hard-to-imitate competitive
manoeuvre through strategic actions to improve their capabilities edge if they are to implement processes of urban regeneration
and competitive edge is deeply rooted. Kitson, Martin, and Tyler and revitalisation successfully. Thus, the resources of a city and
(2004: 991), state that . . .policy has raced ahead of conceptual its ability to manage, develop and maintain those resources give
understanding and empirical analysis, and Turok (2004: 1076), uses rise to a sustainable competitive edge that can be defended against
the term institutionalised competition, based on the idea that it is competitors. The fact that cities compete with one another at
governments and public institutions which are chiey responsible national and/or international level does not mean that they cannot
for improving competitiveness in their regions and cities. Thus, in cooperate or even form partnerships to tackle particular chal-
Europe, competitiveness has been identied as a prime objective in lenges. However, such cooperative arrangements are ultimately
regional policy, and is considered the most important means of intended to make them more competitive (Borja & Castells,
promoting balanced development and territorial cohesion 1997), so they cooperate in order to compete better. This is known
(Enyedi, 2000; Hall, Smith, & Tsoukalis, 2001). But there are also as coopetition (Brandenburger & Nalebuff, 1996).
dissenting voices such as that of Krugman (1996, 1998), who holds Following this explanation of why the concept of competitive-
that it is not nations (or, by extension, regions or cities) that com- ness is applicable to cities, the next step is to explain the Urban
pete with one another but rather businesses, understanding that Competitiveness Index (UCI), the synthetic index drawn up for
actions which only makes sense at microeconomic (business) scale benchmarking cities. It uses a weighted aggregate of the
have been transferred to the macroeconomic (territory) scale. sub-indicators that represent its various dimensions.
78 L. Sez, I. Periez / Cities 48 (2015) 7685
3. The UCI for the location of businesses and attracting of observable indicators (correlations) so that it is possible by grouping
investment individual indicators in terms of sub-indicators to check whether
they are good choices for describing the phenomenon in question.
3.1. Urban competitiveness factors We thus apply factor analysis, more specically principal com-
ponent analysis (PCA), to examine how the different dimensions
Competitiveness, as applied to cities, is a complex, multidimen-
that make up each compound indicator are linked statistically.
sional concept which is hard to measure. It is made up of numerous
Each dimension is a composite designed to describe a particular
underlying factors, some tangible and others intangible, concerned
aspect of the latent phenomenon to be measured (the level of
with inputs and outputs. Authors such as Porter (1991), Begg urban competitiveness in our case). As these aspects are not
(1999), Martin (2003), Turok, Bailey, and Atkinson (2004), directly observable, they are measured by a set of observable indi-
Gardiner et al. (2004), Lengyel (2004), Schwab and Porter (2008), cators which, by denition, are related to the aspect they are sup-
Schwab (2009) and Ferrara and Nistic (2014), and organisations posed to describe and, consequently, to each other. In an ideal
such as the Institute for Management and Development (IMD) situation, each dimension would show a unique, most relevant
(2008), European Union-Regional Policy (2011) via the Joint PCA component accounting for a large degree of the variability
Research Centre (JRC) and the World Economic Forum (2012), con- associated with the full set of indicators. Moreover, all the indica-
sider that competitiveness can be applied to territories in terms of tors should contribute roughly to the same extent and with the
their attractiveness as locations for businesses and for investment same orientation to the most relevant component. PCA is applied
from a multi-dimensional perspective that combines input and to check the internal consistency of each Urban Competitiveness
output factors. Although they do not agree entirely as to what fac- Index dimension. This allows us to detect non-inuencing indica-
tors actually determine, help and improve levels of urban compet- tors, or indicators describing something else or something more
itiveness, they do coincide in categorising them under such aspects than they were supposed to. To begin the analysis we gauge the
as social capital, human capital, quality-of-life, infrastructure, pro- suitability of the sample on each of the three dimensions consid-
ductivity and public institutions that favour the workings of mar- ered by using the KMO (KaiserMeyerOklin) indicator, which
kets. One point on which there is more and more agreement is in shows whether the observable indicators are jointly correlated or
identifying intangible factors such as knowledge, innovation, cre- not by making a comparison of their overall signicance on a scale
ativeness, human capital, the distinctive identity of a city and, from 0 to 1. This comparison is conducted using Bartletts statisti-
more recently, city branding (Ashworth & Kavaratzis, 2010; cal test, which gives statistically signicant gures in all three
Kavaratzis & Ashworth, 2007; Moilanen & Rainisto, 2009; Sez, cases (Table 2). This means that the indicators proposed for each
Periez, & Mediano, 2013; Zhang & Zhao, 2009), as a source of dis- dimension are suitable for obtaining the factors that make up each
tinctiveness and a means of creating sustainable competitive sub-indicator of the UCI. The main hypothesis is thus accepted.
advantage that can be defended over time and against competitors.
Following on from the publications mentioned, our study distin- 3.3. Constructing the UCI
guishes between three dimensions of competitiveness (Table 1):
The basic dimension (B), The efciency dimension (E) and The innovation A synthetic index or compound indicator is a mathematical func-
dimension (I). These dimensions are made up of different competi- tion that associates each element or unit of analysis with a number
tiveness factors that cover both inputs and outputs and affect cities that depends on the characteristics of the element in question, and
differently depending on their level of economic development. The which results in a hierarchical ordering of elements. Thus, if each
basic dimension covers those factors which are understood to be nec- element i is dened by a vector of components X i xi1 ; xi2 ; ; xin
essary for the operation of any urban economy. Their importance in that describes its characteristics then the indicator is a function f
determining the level of competitiveness of a city decreases as its that allocates to each element i a number Ii , according to the values
level of economic development increases. The efciency dimension of the associated vector. A synthetic indicator is required to have
covers factors that describe a more developed, more distinctive urban reasonable properties that make its application plausible, such as
economy in which higher education becomes especially important monotony, scale invariance, translation invariance, continuity and
and the jobs available require a more highly qualied workforce. independence of irrelevant alternatives. It must be pointed out that
The importance of the factors that make up this dimension increases no synthetic indicator can be found that simultaneously meets the
as the level of economic development of the city increases. Finally, requirements of scale and translation invariance, and given that our
the innovation dimension covers factors concerned with generating unit of analysis is cities and that non-homogenous characteristics
knowledge and with ICTs. Cities need to adapt to rapid, complex tech- are assessed, scale invariance is considered more suitable for cata-
nological changes by incorporating new technologies into their man- loguing their relative importance, while translation invariance
agement and production processes. The most powerful cities in would be more suitable for classifying them in absolute terms.
regard to this dimension are those which are most competitive in The synthetic indices or compound indicators used here are
terms of attracting businesses and investments. based on the Nash product (Nash, 1950), which is taken from
cooperative game theory. It consists of allocating to each city the
3.2. The multidimensional structure of urban competitiveness products of the components of its vector of characteristics.
Formally: N x1 ; x2 ; ; xn x1 x2 xn . This indicator weights all
To study the multidimensional nature of urban competitiveness
the components or characteristics of a city in the same way.
we put forward the following main hypothesis: Urban competitive-
Alternatively, characteristics could be weighted in terms of relative
ness in terms of attractiveness for the location of businesses and invest-
weight if the analyst deems it necessary (Freudenberg, 2003). This
ment is a multidimensional concept made up of dimensions that cover
makes it possible to consider alternative synthetic indicators result-
basic, efciency-related and innovation-related factors. On that basis
ing from the weighting of components by means of a system of rel-
we set out three null sub-hypotheses, one for each dimension.1 To x
test these hypotheses, we analyse the relationships between ative weights x xk nk1 ; N x1 ; x2 ; ; xn xx x2 xn
1 x2 xn . In
1
13.000
Basic Sub-indicator
12.000
Although there is a big gap between the top five cities and the
11.000
10.000 rest, it can be seen that the drop is smooth and gradual. In terms
9.000
of the basic dimension, which covers the factors necessary for
Score
8.000
7.000 the operation of any urban economy independent of its level of
6.000
5.000
economic development, the differences between cities seem
4.000 unlikely to be a determinant when it comes to improving or
3.000
guaranteeing a good competitive position in a global setting.
11
21
31
41
51
61
71
81
91
101
111
121
131
141
151
1
Rank
13.000
Eciency Sub-indicator
12.000
A steep downward slope enables five groups of cities to be
11.000
10.000 distinguished in line with their level of competitiveness on the
9.000
efficiency dimension. The differences between them reveal clear
Score
8.000
7.000 disparities in their abilities to compete on this dimension. The
6.000
5.000
factors considered under the efficiency dimension can be
4.000 understood to define the level of urban competitiveness in terms
3.000
of locating businesses and attracting investments.
10
19
28
37
46
55
64
73
82
91
100
109
118
127
136
145
154
1
Rank
13.000
Innovaon Sub-indicator A steep downward slope enables six groups of cities to be
12.000
11.000
10.000
distinguished in line with their level of competitiveness on the
9.000 innovation dimension. The differences between them reveal clear
Score
8.000
7.000 disparities in their abilities to compete on this dimension The
6.000
5.000
factors considered under the efficiency dimension can be
4.000
understood to define the level of urban competitiveness in terms
3.000
10
19
28
37
46
55
64
73
82
91
100
109
118
127
136
145
154
Rank
13.000 are in the top 10, but cities such as Las Palmas de Gran Canaria
Very High
Urban Compeveness Score
12.000 and Cordoba in Spain and Tarento and Caserta in Italy are at the
High
11.000 bottom of the ranking. This highlights the great disparities that
Medium
10.000
Low exist in terms of attractiveness and urban competitiveness in these
9.000
countries. The cities with the worst results are mainly located in
8.000
7.000
Eastern Europe; the only Eastern European cities to obtain good
6.000 results are national capitals, indicating the extent to which factors
5.000 for competition are concentrated in them.
4.000
3.000
4.2. Correlation between the UCI and its dimensions
103
109
115
121
127
133
139
145
151
157
13
19
25
31
37
43
49
55
61
67
73
79
85
91
97
1
7
Table 3
Correlation matrix between UCI and its sub-indicators.
The bold values show that there is moderate correlation between the UCI and the basic sub-indicator (0.465), but the correlation with the
efciency indicator (0.970) and with the innovation indicator (0.973) is high. Bold values (signicance) show that those correlations are indeed
statistically signicant.
Table 4
Correlation matrix between UCI, sub-indicators and PPP per capita GDP 2009.
The bold values show that the correlation between the three sub-indicators and PPP per capita GDP reveals that, with the exception of the Basic sub-indicator, for which the
gure is just 0.459 (though this is still statistically signicant), high correlation levels of more than 0.7 are obtained. And there is statistically signicant correlation between
the UCI and PPP per capita GDP (0.740).
PPP per capita GDP (0.740). This means that the sub-indicators
used are, in general, closely related to PPP per capita GDP and that sub-indicators obtained, a second test was run that comprises
the method used for constructing the compound indicator or syn- varying the weights initially allocated according to the levels of
thetic index is sound. An examination of the correlation between development of each city for each of the dimensions examined
the three sub-indicators and PPP per capita GDP reveals that, with (basic, efciency-related and innovation-related) (Table 53).
the exception of the Basic sub-indicator, for which the gure is just
0.459 (though this is still statistically signicant), high correlation 3
The range of variation allocated to the weights was obtained using Excel, which is
levels of more than 0.7 are obtained. capable of generating pseudo-random numbers from a uniform distribution between
To obtain a more detailed picture of how robust the UCI is, and zero and one with the function = a + (b a) RANDOM (), where a = bottom end and
thus conrm the suitability of its construction based on the three b = top end.
82 L. Sez, I. Periez / Cities 48 (2015) 7685
M/I/Ha Code City Score M/I/H Code City Score M/I/H Code City Score M/I/H Code City Score
H UK001L London 10,749 H DE012L Bremen 8,424 H UK016L Aberdeen 7,808 M IT008L Bari 7,307
H FR001L Paris 10,747 H NL001L The Hague 8,417 H NL007L Groningen 7,803 M IT010L Catania 7,305
H ES001L Madrid 9,847 H IT009L Bologna 8,409 H NL010L Heerlen 7,795 I UK026L Kingston 7,250
H IT002L Milan 9,710 I HU001L Budapest 8,407 I UK009L Cardiff 7,773 I DE018L Halle 7,226
H ES002L Barcelona 9,616 H BE002L Antwerp 8,400 H ES014L Pamplona 7,757 I DE021L Gottingen 7,214
H IT001L Roma 9,567 H DE017L Bielefeld 8,399 I DE009L Dresden 7,739 M IT005L Palermo 7,211
H DE003L Munich 9,532 H UK007L Edinburgh 8,345 H DE025L Darmstadt 7,732 I EE001L Tallinn 7,200
H EL001L Athens 9,388 H SE003L Malmo 8,339 H DK004L Aalborg 7,728 M PT002L Oporto 7,092
H DE005L Frankfurt 9,384 H IT007L Florence 8,334 I EL002L Thessalonica 7,709 I BE004L Charleroi 7,065
H DE002L Hamburg 9,325 H LU001L Luxemburg 8,328 I UK027L Stoke-on-trent 7,707 M PL005L Poznan 6,989
H DE007L Stuttgart 9,222 H ES019L Bilbao 8,323 H AT002L Graz 7,701 M PL003L Krakow 6,960
H AT001L Vienna 9,177 H IT028L Padua 8,268 I DE008L Leipzig 7,694 M IT021L Caserta 6,915
H SE001L Stockholm 9,152 H FR013L Rennes 8,215 H AT004L Salzburg 7,656 M BG001L Sopia 6,914
H BE001L Brussels 9,098 H IT012L Verona 8,147 H FR020L Dijon 7,654 I UK022L Wrexham 6,854