3rd Central European Conference in Regional Science CERS, 2009 45

Smart cities in Europe

ANDREA CARAGLIU
Politecnico di Milano, Piazza Leonardo 32, 20133 Milan, Italy.

CHIARA DEL BO
Universit degli Studi di Milano, Via Conservatorio 7, 20122 Milan, Italy.

PETER NIJKAMP
VU University, De Boelelaan 1105, Amsterdam 1081 HV, The Netherlands

Abstract
Urban performance currently depends not only on the citys endowment of hard infrastructure (physical
capital), but also, and increasingly so, on the availability and quality of knowledge communication and social
infrastructure (human and social capital). The latter form of capital is decisive for urban competitiveness.
Against this background, the concept of the smart city has recently been introduced as a strategic device to
encompass modern urban production factors in a common framework and, in particular, to highlight the
importance of Information and Communication Technologies (ICTs) in the last 20 years for enhancing the
competitive profile of a city.
The present paper aims to shed light on the often elusive definition of the concept of the smart city. We provide
a focussed and operational definition of this construct and present consistent evidence on the geography of
smart cities in the EU27. Our statistical and graphical analyses exploit in depth, for the first time to our
knowledge, the most recent version of the Urban Audit data set in order to analyse the factors determining the
performance of smart cities.
We find that the presence of a creative class, the quality of and dedicated attention to the urban environment, the
level of education, multimodal accessibility, and the use of ICTs for public administration are all positively
correlated with urban wealth. This result prompts the formulation of a new strategic agenda for smart cities in
Europe, in order to achieve sustainable urban development and a better urban landscape.

Keywords: smart city, urban development, human capital, transport infrastructure, ICTs

JEL classification codes: A13, L90, O18, R12

1. Introduction

What is the source of urban growth and of sustainable urban development? This question has received
continuous attention from researchers and policy makers for many decades. Cities all over the world
are in a state of flux and exhibit complex dynamics. As cities grow, planners devise complex systems
to deal with food supplies on an international scale, water supplies over long distances and local
waste disposal, urban traffic management systems and so on; () and the quality of all such urban
inputs defines the quality of life of urban dwellers (The Science Museum 2004).
 3rd Central European Conference in Regional Science CERS, 2009 46

Notwithstanding the enormous formidable challenges and disadvantages associated with urban
agglomerations, the world population has been steadily concentrating in cities. Figure 1 shows the
percentage of US citizens living in cities (defined as agglomerations of more than 1,000 dwellers); a
massive rise in this percentage took place, from 5.1 per cent in 1790 to more than 75 per cent of the
US population being located in urban areas in the year 2000.
80
% of US population living in cities
20 40 0 60

1800 1850 1900 1950 2000

Year

Figure 1 Percentage of US population living in urban areas, 1790-1990

Source: US Census
In addition, we also witness a substantial increase in the average size of urban areas. This has been
made possible by a simultaneous upward shift in the urban technological frontier, so that a city could
accommodate more inhabitants. Problems associated with urban agglomerations have usually been
solved by means of creativity, human capital, cooperation (sometimes bargaining) among relevant
stakeholders, and bright scientific ideas: in a nutshell, smart solutions. The label smart city should
therefore point to clever solutions allowing modern cities to thrive, through quantitative and
qualitative improvement in productivity. However, when googling Smart city definition1, we
discover that among the very first results we can name a communications provider, a US radio, an
Edinburgh hostel, an initiative of the Amsterdam Innovation Engine, and so on; but no sign of a proper
definition.

In the present paper we search for a clearer and focussed definition of the label smart city. We next
provide qualitative evidence on the correlations between the dimensions of our definition of smart
cities and a measure of wealth, i.e. per capita GDP in Purchasing Power Parity (henceforth, PPP).2 We
will start with a brief literature review in the next section.

1
This Google search has been carried out on 8 April 2009.
2
PPP methods make it possible to better represent spatial disparities in the level of prices, and, consequently,
more accurately gauge the real spending power of economic agents.
 3rd Central European Conference in Regional Science CERS, 2009 47

2. Literature review
The concept of the smart city has been quite fashionable in the policy arena in recent years. Its main
focus seems to be on the role of ICT infrastructure, although much research has also been carried out
on the role of human capital/education, social and relational capital and environmental interest as
important drivers of urban growth.

The European Union (EU), in particular, has devoted constant efforts to devising a strategy for
achieving urban growth in a smart sense for its metropolitan areas. Not only the EU, but also other
international institutions and thinktanks believe in a wired, ICT-driven form of development. The
Intelligent Community Forum produces, for instance, research on the local effects of the ICT
revolution, which is now available worldwide. The OECD and EUROSTAT Oslo Manual (2005)
stresses instead the role of innovation in ICT sectors and provides a toolkit to identify consistent
indicators, thus shaping a sound framework of analysis for researchers on urban innovation. At a
meso-regional level, we observe renewed attention for the role of soft communication infrastructure in
determining economic performance.3

The availability and quality of the ICT infrastructure is not the only definition of a smart or intelligent
city. Other definitions stress the role of human capital and education in urban development. Berry and
Glaeser (2005) and Glaeser and Berry (2006) show, for example, that the most rapid urban growth
rates have been achieved in cities where a high share of educated labour force is available. In
particular Berry and Glaeser (2005) model the relation between human capital and urban development
by assuming that innovation is driven by entrepreneurs who innovate in industries and products which
require an increasingly more skilled labour force. As not all cities are equally successful in investing
in human capital, the data show that an educated labour force or, in Floridas jargon, the creative
class is spatially clustering over time. This recognized tendency of cities to diverge in terms of
human capital levels has attracted the attention of researchers and policy makers. It turns out that some
cities, which were in the past better endowed with a skilled labour force, have managed to attract more
skilled labour, whereas competing cities failed to do so. Policy makers, and in particular European
ones, are most likely to attach a consistent weight to spatial homogeneity; in these circumstances the
progressive clusterization of urban human capital is then a major concern.

The label smart city is still, in our opinion, quite a fuzzy concept. We can summarize the
characteristics proper to a smart city that tend to be common to many of the previous findings as
follows:4

1. The utilization of networked infrastructure to improve economic and political efficiency and
enable social, cultural and urban development5, where the term infrastructure indicates business
services, housing, leisure and lifestyle services, and ICTs (mobile and fixed phones, satellite TVs,
computer networks, e-commerce, internet services). This point brings to the forefront the idea of a
wired city as the main development model and of connectivity as the source of growth.

3
Del Bo and Florio (2008) offer a critical perspective on previous studies regarding the role of different forms of
infrastructure in economic performance and provide empirical evidence on the contribution of single and
aggregate measures of infrastructure on regional growth in the period 1995-2005.
4
This section summarizes and further elaborates the main points in Hollands (2008), adding a critical review of
the literature on urban growth from an economists perspective.
5
The use of italics in this list indicates a citation from Hollands (2008). On this first point, see also Komninos
(2002).
 3rd Central European Conference in Regional Science CERS, 2009 48

2. An underlying emphasis on business-led urban development. According to several critiques of

the concept of the smart city, this idea of neo-liberal urban spaces, where business-friendly cities
would aim to attract new businesses, would be misleading. However, although caveats on the
potential risks associated with putting an excessive weight on economic values as the sole driver
of urban development may be worth noting the data actually show that business-oriented cities are
indeed among those with a satisfactory socio-economic performance.

3. A strong focus on the aim to achieve the social inclusion of various urban residents in public
services (e.g. Southamptons smartcard).6 This prompts researchers and policy makers to give
attention to the crucial issue of equitable urban growth. In other words: To what extent do all
social classes benefit from a technological impulse to their urban fabric?

4. A stress on the crucial role of high-tech and creative industries in long-run urban growth. This
factor, along with soft infrastructure (knowledge networks, voluntary organizations, crime-free
environments, after dark entertainment economy), is the core of Richard Floridas research.7 The
basic idea in this case is that creative occupations are growing and firms now orient themselves
to attract the creative. Employers now prod their hires onto greater bursts of inspiration. The
urban lesson of Floridas book is that cities that want to succeed must aim at attracting the
creative types who are, Florida argues, the wave of the future (Glaeser 2005). The role of
creative cultures in cities is also critically summarized in Nijkamp (2008), where creative capital
co-determines, fosters and reinforces trends of skilled migration. While the presence of a creative
and skilled workforce does not guarantee urban performance, in a knowledge-intensive, and
increasingly, globalized economy, these factors will determine increasingly the success of cities.

5. Profound attention to the role of social and relational capital in urban development. A smart city
will be a city whose community has learned to learn, adapt and innovate (Coe et al 2001). People
need to be able to use the technology in order to benefit from it: this refers to the absorptive
capacity literature.8 When social and relational issues are not properly taken into account, social
polarization may arise as a result. This last issue is also linked to economic, spatial and cultural
polarization. It should be noted, however, that some research actually argues the contrary.
Poelhekke (2006), for example, shows that the concentration of high skilled workers is conducive
to urban growth, irrespective of the polarization effects that this process may generate at a meso-
(for example, regional) level. The debate on the possible class inequality effects of policies
oriented towards creating smart cities is, however, still not resolved.

6. Finally, social and environmental sustainability as a major strategic component of smart cities. In
a world where resources are scarce, and where cities are increasingly basing their development
and wealth on tourism and natural resources, their exploitation must guarantee the safe and
renewable use of natural heritage. This last point is linked to the third item, because the wise
balance of growth-enhancing measures, on the one hand, and the protection of weak links, on the
other, is a cornerstone for sustainable urban development.

6
See Southampton City Council 2006.
7
See, e.g., Florida (2002).
8
This concept has been applied to different economic relations at different levels of spatial aggregation. The
basic reference is Cohen and Levinthal (1990); Abreu et al. (2008) bridges the idea from a micro-, firm level to a
more aggregated, meso-level; finally, Caragliu and Nijkamp (2008) test the role of regional absorptive capacity
in inducing spatial knowledge spillovers.
 3rd Central European Conference in Regional Science CERS, 2009 49

Items 5 and 6 are for us the most interesting and promising ones, from both a research and a policy
perspective. In the next sections we provide quantitative and analytical evidence on the role of the
creative class and human capital in sustainable urban development, arguing that it is indeed the mix of
these two dimensions that determine the very notion of a smart city. The relational capital side of the
story is not evaluated in the present paper, but this will be the subject of further research in future
studies.

Along with the previously mentioned critical points, additional critiques have been advanced to
question the concept of a smart or intelligent city. Hollands (2008) provides a thorough treatment of
the main arguments against the superficial use of this concept in the policy arena. His main points are
the following:

The focus of the concept of smart city may lead to an underestimation of the possible negative
effects of the development of the new technological and networked infrastructures needed for
a city to be smart (on this topic, see also Graham and Marvin 2001);

This bias in strategic interest may lead to ignoring alternative avenues of promising urban
development;

Among these possible development patterns, policy makers would better consider those that
depend not only on a business-led model. As a globalized business model is based on capital
mobility, following a business-oriented model may result in a losing long term strategy: The
spatial fix inevitably means that mobile capital can often write its own deals to come to
town, only to move on when it receives a better deal elsewhere. This is no less true for the
smart city than it was for the industrial, manufacturing city.9
Our paper will now provide some quantitative evidence on these points, supported by spatial statistics,
maps and graphical evidence on each of the points that the literature on smart cities has put forward, in
order to explore and identify statistical correlations with socio-economic urban performance.

3. An operational definition of the smart city

A narrow definition of a much-used concept may help in understanding the scope of the present paper.
Although several different definitions of smart city have been given in the past, most of them focus on
the role of communication infrastructure. However, this bias reflects the time period when the smart
city label gained interest, viz. the early 1990s, when the ICTs first reached a wide audience in
European countries. Hence, in our opinion, the stress on the internet as the smart city identifier no
longer suffices.

A recent and interesting project conducted by the Centre of Regional Science at the Vienna University
of Technology identifies six main axes (dimensions) along which a ranking of 70 European middle
size cities can be made. These axes are: a smart economy; smart mobility; a smart environment; smart
people; smart living; and, finally, smart governance. These six axes connect with traditional regional
and neoclassical theories of urban growth and development. In particular, the axes are based
respectively on theories of regional competitiveness, transport and ICT economics, natural
resources, human and social capital, quality of life, and participation of societies in cities. We believe

9
Hollands (2008), p. 314.
 3rd Central European Conference in Regional Science CERS, 2009 50

this offers a solid background for our theoretical framework, and therefore we base our definition on
these six axes.

We believe a city to be smart when investments in human and social capital and traditional
(transport) and modern (ICT) communication infrastructure fuel sustainable economic growth and a
high quality of life, with a wise management of natural resources, through participatory governance.
 3rd Central European Conference in Regional Science CERS, 2009 51

4. Quantitative and graphical evidence on European smart cities

In this section we will present graphical and quantitative evidence on the relative performance and
rankings of European cities with respect to measures reflecting some of the definitions of a smart city
given in the literature. The data source is the Urban Audit data set in its latest wave (2003-2006).10
Cities that were surveyed in the latest available wave are depicted in Map 1.

Troms

Oulu
Trondheim Ume

Tampere
Bergen Turku
Oslo Uppsala
rebro Stockholm Tallinn
Stavanger
Tartu
Gteborg Linkping
Aberdeen Aalborg Jnkping Riga
Aarhus Liepaja
Glasgow Malm
Derry Odense
Edinburgh Kaunas
Kiel KbenhavnKoszalinGdansk Suwalki
BelfastWirralLeeds Bremen
Lincoln Alkmaar Olsztyn Bialystok
Almere MagdeburgBerlinPoznanKonin
WrexhamLondon Lodz
Exeter Gent Kln Erfurt LiberecOpole Radom
Lille
Bonn Mainz WeimarPlzen Zory Kielce
Portsmouth
Caen Paris Metz Brno
Kosice
Rennes Augsburg MnchenLinz Zlin
Tours Dijon Gyor Nitra Miskolc
Nantes Bern Salzburg Graz
Poitiers
Limoges Lyon Trento Pecs Arad Sibiu Bacau
Torino Padova Maribor Szeged Braila
Bordeaux Verona
Toulouse Nice Vidin Ruse
OviedoGijn Bilbao Toulon Modena Varna
Santiago de Compostela Sofia Burgas
Vigo SantanderLogroo Montpellier Marseille Ajaccio Roma
Braga Zaragoza
Aveiro Madrid Barcelona Sassari Napoli Bari ThessalonikiKavala
Toledo Caserta Taranto
Coimbra Valencia Palma di MallorcaCagliari Potenza LarisaVolos
Lisboa Ioannina
Crdoba Murcia Palermo Patra Athina
SetbalFaro Catania Kalamata
Mlaga
Gozo
Valletta Irakleio Lefkosia

Funchal

Las Palmas

Map 1: Cities in the 2003-2006 Urban Audit survey

10
The Urban Audit entails a collection of comparable statistics and indicators for European cities; it contains
data for over 250 indicators across the following domains:
Demography;
Social aspects;
Economic aspects;
Civic involvement;
Training and education;
Environment;
Travel and transport;
Information society;
Culture and recreation.
 3rd Central European Conference in Regional Science CERS, 2009 52

We now present a set of charts which show partial correlations between urban growth determinants
and our measure of economic output, which is per capita GDP in purchasing power standards (PPS) in
2004 (the latest data available in the Urban Audit data set).

The set of all partial correlations among the variables we use to measure the smartness of European
cities can be found in Table 1, with corresponding p-values in parentheses. It is evident that most of
the variables which we deem as capable of both co-determining long-run urban performance and
characterizing a thorough definition of smart city, tend to be positively associated with our measure of
urban wealth (we chose per capita GDP in PPS in 2004 in order to avoid the problem of size effects
and to take into account price differentials across countries, which might be particularly different
among EU15 and New Member State (NMS) cities).11 Throughout this section, on the map as well as
in our charts, we indicate the name of the city associated with each observation. We believe this to be
a useful tool of analysis for both researchers as well as policymakers, to identify intriguing spatial
issues in the Urban Audit data set, the possible presence of country effects, and more in general to
allow the reader to identify the locational patterns of our smart city measures.

Table 1 Partial correlations between the sic indicators of Smart Cities

Employment in Length of public
Per capita GDP Multimodal e- Human
the entertainment transport
in PPS accessbiility Government capital
industry network
Per capita GDP in
1
PPS
Employment in 0.215 1
the entertainment
industry (0.1258)
Multimodal 0.7049 -0.0059 1
accessibility 0 (0.9553)
Length of public 0.3104 0.2874 0.0919 1
transport network (0.0043) (0.0302) (0.312)
0.1418 -0.0254 0.141 -0.0339 1
e-Government
(0.1751) (0.8385) (0.1004) (0.7417)
-0.1361 -0.0983 0.0833 -0.0741 0.0665 1
Human capital
(0.265) (0.3649) (0.3616) (0.5946) (0.5733)
Note: p-values are in parentheses

Figure 2 offers partial support for Richard Floridas arguments on the role of the creative class in
determining long-run urban performance. Positive correlations between the share of people employed
in a creative industry12, and in particular in the super-creative core13, are found in US cities and

11
An interesting but puzzling result arises for the relationship between the level of education of people living in
our sample and their average individual income; this issue will be further analysed later in this section.
12
See Florida (2002, 2009).
13
In Florida (2002) the creative class is defined as the merger of two Standard Occupational Classification
System codes within the US labour force, viz.:
A super-creative core with those employed in science, engineering, education, computer programming,
research, and with arts, design, and media workers making a small subset. Those belonging to this group are
considered to fully engage in the creative process (Florida, 2002, p.69);
 3rd Central European Conference in Regional Science CERS, 2009 53

states. Here, we measure these effects with the share of the labour force in European cities in the
culture and entertainment industry, and find indeed that the two measures show a positive and
significant correlation (the correlation coefficient equals .2150 with a p-value of .1258).

In the urban economics literature, Floridas view has not been exempt from criticism.14 In the opinion
of several economists, the argument that the creative professions would drive urban performance is
flawed, and it would only be a proxy for the role of the hard measurable stock of human capital (i.e.
technical professions and total years of schooling) on urban growth. Shapiro (2008) provides an
excellent and convincing bridge between the two views. In his paper he proves with careful
econometric estimations that human capital in cities contributes both directly to urban growth
(measured by the growth of population, wages and two land rent measures) through productivity gains
and indirectly through the increase in urban amenities, which in turn may foster the process of
attraction of the creative class. Although the productivity effects are still the largest, according to
Shapiros estimates the amenities effects would account for as much as 20 to 30 per cent of total
human capital effects on urban growth.15
800

Frankfurt am Main
600
GDP per head in PPS in 2004

RegensburgDsseldorf

Stockholm
Amsterdam Mnchen
Darmstadt
Karlsruhe
400

Hamburg
Wiesbaden
Nrnberg Mainz
Hannover Augsburg Kln
BremenGteborgMadrid
Trier Bonn
Essen FreiburgBudapest Malm Bratislava
im Breisgau
Dresden
Jnkping
Bochum Bielefeld
Ume (Oder)
Schwerin
Mlheim a.d.RuhrErfurt
Dortmund
Wuppertal Frankfurt
Magdeburg
Murcia
Mnchengladbach
200

Berlin Leipzig
Halle an der Saale Tallinn
Lefkosia Weimar

Kosice
Nitra Tartu
Banska
Pecs Bystrica
Miskolc
Nyiregyhaza
0

0 2 4 6
Proportion of employment in culture and entertainment industry in 2004

Figure 2 Creative class and wealth in 2004

Creative professionals with those employed in healthcare, business and finance, the legal sector, and
education.
14
See, for example, Glaeser (2005).
15
The direct causal mechanism will be graphically analysed later in this section.
 3rd Central European Conference in Regional Science CERS, 2009 54

A second positive (and extremely significant) correlation appears to exist between multimodal
accessibility and per capita GDP (Figure 3). In this chart, the accessibility indicator, calculated as a
weighted average of the ease with which a city can be reached with a combined set of available
transportation modes (i.e. rail, road, sea or plane), also represents a measure for the market potential
available to and from the city itself. Therefore, a better endowment of transportation means might be
conducive to wealth and growth, this last statement being in line with the New Economic Geographys
theoretical expectations.16
800

Frankfurt am Main
600
GDP per head in PPS in 2004

Regensburg Dsseldorf

Bruxelles
Stockholm
Mnchen Amsterdam
Darmstadt
Copenhagen
Karlsruhe Hamburg
400

Wiesbaden
Augsburg Hannover Kln Mainz
Nrnberg
Pamplona/Irua Gteborg Madrid Bremen
Vitoria/Gasteiz Trier
Barcelona Bonn
Antwerpen
Malm
Warszawa
Bratislava Essen
Aarhus Palma di MallorcaBudapest
Freiburg im
Gent Breisgau
Valladolid Lisboa
Aalborg
Jnkping Dresden
Logroo Ume SantanderSchwerin
Odense Bielefeld
Erfurt Bochum
Dortmund Mlheim a.d.Ruhr
Sevilla Valencia
Oviedo Brugge
Frankfurt (Oder)
MurciaZaragoza Magdeburg
Poznan Wuppertal
Mnchengladbach
200

Badajoz Toledo Tallinn Leipzig

HalleLigeder Saale Berlin
anCharleroi
Coimbra Riga
Mlaga Weimar Wien
Lefkosia Oporto
Gdansk Krakow
Aveiro Wroclaw
Tartu Lodz
Braga
Setubal Katowice
Liepaja Pecs Banska
Bydgoszcz Kosice
Nitra
Zory
Bystrica
Szczecin
Olsztyn
Torun
Bialystok
Suwalki Zielona
Gorzow
Miskolc
Opole
Konin
Rzeszow Gora
Wielkopolski
Jelenia Gora
Lublin
Kielce
Nyiregyhaza
Nowy Sacz
0

0 50 100 150 200

Multimodal accessibility (EU27=100) in 2004

Figure 3 Accessibility and wealth in 2004

Figure 4 shows instead the relationship between the availability of public transportation (normalized
by the city area) and the level of wealth, measured as before with per capita GDP in PPS. The
relationship is strongly positive; the city of Stockholm has been excluded from the original dataset as
it behaves as an outlier, with an outstandingly high density of public transportation. With the inclusion
of Stockholm the interpolation line would become even steeper. It is quite evident that an efficient net
of public transportation is associated with high levels of wealth. Although the direction of causality in
this relation may go both ways, it seems reasonable to think that a dense public transportation network
may help to reverse the negative effects of urban density, thus at least partly releasing the pressure this
exerts on the urban landscape and reducing the costs associated with congestion.

16
For the role of the market potential in driving economic performance in the New Economic Geography
literature, we refer to Redding and Sturm (2008), amongst others.
 3rd Central European Conference in Regional Science CERS, 2009 55

800

Frankfurt am Main
600
GDP per head in PPS in 2004

Regensburg Dsseldorf

Bruxelles
Mnchen
Darmstadt
KarlsruheHamburg
400

Wiesbaden Mainz Nrnberg

Augsburg
Kln Hannover
Bremen Gteborg Madrid
Trier Bonn
Malm Bratislava
Warszawa
Freiburg im Breisgau Essen
Budapest Aarhus
Valladolid
Bielefeld
Schwerin Bochum
Brugge a.d.Ruhr Wuppertal
Frankfurt (Oder)
Mlheim
Magdeburg Poznan
200

Funchal Lige Halle anLeipzig

derTallinn
Saale
Weimar
Coimbra Wien Riga
Krakow
Gdansk Wroclaw Oporto
Aveiro
Braga Lodz Katowice
Tartu
Setubal
Nitra
Banska
Szczecin Zory
Bystrica
Bydgoszcz
Zielona
Gorzow
Olsztyn
Opole Gora
Liepaja
Wielkopolski PecsKosice
JeleniaTorun
KoninGora
Bialystok
Rzeszow
Kielce
Nyiregyhaza Suwalki
Lublin
Nowy Sacz
0

0 200 400 600 800

Length of public transport network / land area in 2004

Figure 4 Public transport and wealth

A slightly less significant and less steep association can be found between the level of GDP and a
measure of e-government. The Urban Audit data set yields both the absolute number of government
forms that can be downloaded from the website of the municipal authority, as well as the number of
administrative forms which can be submitted electronically. As this last series has slightly more
observations, and is, in our opinion, a better measure of the real chance for citizens to interact with the
urban Public Administration via the net, we represent this in Figure 5. The city of Krakow is in this
case excluded as an outlier (in terms of number of forms that can be submitted online). The
relationship does not change when the e-government measure is normalized by population or labour
force (although this operation slightly changes the relative ranking of the cities in our sample).

Although cities with a high level of per capita GDP also tend to devote more attention to smart, e-
government solutions, it is interesting to observe that some noticeable exceptions characterize this
analysis. Some cities in peripheral countries (Krakow in Poland, Zaragoza in Spain, Ponto Delgada in
Portugal) have also devised a wide set of forms that citizens can submit online, thus reducing travel
and commuting costs, and costs associated with the management of multi-task public administration
bodies.
 3rd Central European Conference in Regional Science CERS, 2009 56

Bruxelles
500 Stockholm
Mnchen
Amsterdam
Darmstadt

Karlsruhe
400
GDP per head in PPS in 2004

Wiesbaden
Nrnberg
Kln Augsburg
Gteborg Madrid
Pamplona/Irua Bremen
Bonn Barcelona Antwerpen
Vitoria/Gasteiz
300

Malm
Warszawa
Essen
Bratislava
Budapest
Palma di MallorcaGent
Dresden Aarhus
Lisboa
Valladolid
JnkpingLogroo Aalborg
Bielefeld
Santander
Schwerin
Erfurt Bochum
Brugge
Las Dortmund
Palmas
Frankfurt (Oder)
Valencia
Oviedo
Sevilla Mlheim a.d.Ruhr
Zaragoza
Murcia
Poznan
Mnchengladbach
Leipzig
Tallinn
Berlin
200

Toledo
Halle an der Saale
Badajoz Funchal
Lige
Charleroi
Mlaga
Coimbra
Oporto
Wroclaw
Aveiro Gdansk
Ponto Delgada
Katowice
Lodz Tartu
Braga
Setubal
Kosice
Liepaja
Nitra
Zory
100

Banska Bystrica Bydgoszcz

Pecs Szczecin
Gorzow
Olsztyn
Opole
Torun Zielona Gora
Wielkopolski
Jelenia
Konin
Rzeszow
Suwalki Gora
Bialystok
Lublin
Kielce Nyiregyhaza
Nowy Sacz

0 50 100 150 200

Number of administrative forms which can be submitted electronically in 2004

Figure 5 e-Government and wealth

Finally, Figure 6 shows the relationship between the stock of human capital and the level of urban
wealth. According to neoclassical theories (Lucas 1988, Arrow 1962, Mankiw et al. 1992), human
capital levels are good predictors of subsequent economic performance. As Table 1 shows, in our
sample this positive relationship has, nevertheless, more complex characteristics. The correlation
coefficient between our measure of human capital, i.e. the share of the labour force qualified at ISCED
levels 3 and 4,17 and the level of GDP is negative (although not significant at any statistical confidence
level). Does this imply that more education is associated with poorer economic conditions? If we look
at Figure 5 it seems clear that the correct fit of this relationship is through a quadratic interpolation.
After an appropriate (quadratic) term has been taken into account, the linear correlation between
human capital and GDP is positive and significant at the 1 per cent level.18

The interpretation of this finding is, however, more difficult. By inspecting Figure 5 it is possible to
identify some observations on the right-hand side of the chart as cities in the new Member States of

17
The International Standard Classification of Education (ISCED) was designed by UNESCO in the early
1970s to serve as an instrument suitable for assembling, compiling and presenting statistics of education both
within individual countries and internationally. It was approved by the International Conference on Education
(Geneva, 1975), and was subsequently endorsed by UNESCOs General Conference when it adopted the Revised
Recommendation concerning the International Standardization of Educational Statistics at its twentieth session
(Paris, 1978) (from unesco.org).
18
Evidence of this last finding is available from the authors upon request.
 3rd Central European Conference in Regional Science CERS, 2009 57

the EU. As a legacy of the communist period, when levels of education were deliberately held high,
labour forces in those countries may still own a large stock of human capital, albeit that overall levels
of individual wealth may not yet match those of the old Member States. In this case, therefore, the
depicted
800relationship may actually represent an off-saddle growth path portrait of the real human
capital-urban growth equation.19

Frankfurt am Main
GDP per head in PPS in 2004

600
Dsseldorf
Regensburg

Stockholm
Amsterdam Mnchen
Darmstadt
Copenhagen
400 Karlsruhe Hamburg
Wiesbaden
Mainz
Nrnberg
Kln Hannover
Augsburg
Madrid Pamplona/Irua Gteborg
Bremen
Barcelona
Vitoria/Gasteiz Bonn
Trier Malm Essen
Budapest
Palma di Freiburg im Breisgau
Mallorca Aarhus
Dresden
ValladolidSantander
Logroo UmeAalborg
Jnkping
Erfurt Bielefeld
Schwerin
Odense Bochum
Las Palmas
Oviedo
Sevilla Valencia Dortmund
Frankfurt (Oder)
Mlheim a.d.Ruhr
200 Murcia Zaragoza Wuppertal
MnchengladbachMagdeburg
Toledo Berlin Leipzig Halle an der Saale
Tallinn
Mlaga Weimar Wien Riga
Lefkosia
Tartu Nitra
Miskolc Banska Bystrica
Nyiregyhaza

0
20 30 40 50 60 70
Prop. of working age population qualified at ISCED levels 3 or 4 in 2004

Figure 6 Human capital and wealth

A second key to interpret the puzzle may be by reconnecting our study to Mayer (2007). She analyses
the different ways in which cities and regions can set up a high-technology cluster even without the
presence of a sound research-oriented university, whilst also criticizing the opposite side of the story,
viz. the idea that academic research centres are a necessary and sufficient condition for achieving
high-tech oriented urban development. Therefore, cities in new Member States may still fail to provide
a sound connection between academic research institutes and the real economy, thus failing to attract
the human capital-rich workers who raise productivity and wealth.

19
Indirect evidence to support this guess comes from splitting the sample into countries that in the 1980s were
liberal or capitalist in Europe and those which belonged to COMECON, and then fitting the data with a linear
trend; the latter turns out to be positive and significant for the first of these two subsamples and negative and
significant for the second .
 3rd Central European Conference in Regional Science CERS, 2009 58

5. Conclusions and policy implications

In this paper, we have presented an overview of the concept of the smart city, with a critical review
of the previous economics and planning approaches to this concept. We then presented a narrower
definition of the concept of the smart city, and reviewed some quantitative and graphical evidence on
the correlations of some of the main determinants of economic performance and the most important
measure of urban success, viz. per capita wealth.

Data from the 2004 Urban Audit data set show consistent evidence of a positive association between
urban wealth and the presence of a vast number of creative professionals, a high score in a multimodal
accessibility indicator, the quality of urban transportation networks, the diffusion of ICTs (most
noticeably in the e-government industry), and, finally, the quality of human capital. These positive
associations clearly define a policy agenda for smart cities, although clarity does not necessarily imply
ease of implementation.

All variables shown to be positively associated with urban growth can be conceived of as stocks of
capital; they are accumulated over time and are subject to decay processes. Hence, educating people is
on average successful only when investment in education is carried out over a long period with a
stable flow of resources; transportation networks must be constantly updated to keep up with other
fast-growing cities, in order to keep attracting people and ideas; the fast pace of innovation in the ICT
industry calls for a continuous and deep restructuring and rethinking of the communication
infrastructure, to prevent European cities from losing ground to global competitors.

This continuous challenge, the endless frontier to quote Vannevar Bushs words on scientific
research (Bush 1945), is the only way to ensure a sustainable path of development for cities, whilst at
the same time guaranteeing that cities will maintain their crucial role as the cradle of ideas and
freedom.

References
Abreu, M., Grinevich, Vadim., Kitson M. and Savona M. (2008). Absorptive capacity and regional patterns of
innovation, Research Report DIUS RR-08-11,Cambridge MA: MIT.
Arrow, K. J.(1962). The economic implications of learning by doing, Review of Economic Studies 29, 155-
177.
Berry, C. R. and Glaeser, E.L. (2005). The divergence of human capital levels across cities, Papers in
Regional Science, 84(3), 407-444.
Bush, V. (1945). Science: the endless frontier, Washington DC: United States Government Printing Office.
Caragliu, A. and Nijkamp, P. (2008). The impact of regional absorptive capacity on spatial knowledge
spillovers, Tinbergen Institute Discussion Papers 08-119/3, Amsterdam: Tinbergen Institute.
Coe, A., Paquet, G. and Roy, J. (2001). E-governance and smart communities: a social learning challenge,
Social Science Computer Review, 19 (1), 80-93.
Cohen W., and Levinthal, D. (1990). Absorptive capacity: a new perspective on learning and innovation,
Administrative Science Quarterly, 35 (1), 128-152.
Del Bo, C. and Florio, M. (2008). Infrastructure and growth in the European Union: an empirical analysis at the
regional level in a spatial framework, Departmental Working Papers 2008-37, Milan: University of Milan,
Department of Economics.
Florida, R. L. (2002). The rise of the creative class: and how it's transforming work, leisure, community and
everyday life, New York: Basic Books.
 3rd Central European Conference in Regional Science CERS, 2009 59

Florida, R. L. (2009). Class and Well-Being, retrieved on the net on 17 March 2009,7:38am EDT.
http://www.creativeclass.com/creative_class/2009/03/17/class-and-well-being/
Giffinger, R, Fertner, C., Kramar, H., Kalasek, R. Pichler-Milanovi, N., and Meijers, E. (2007). Smart cities -
Ranking of European medium-sized cities, Final report (October 2007). On the web: http://www.smart-
cities.eu/download/smart_cities_final_report.pdf
Glaeser, E.L. (2005). A review of Richard Floridas The rise of the creative class, Regional Science and
Urban Economics, 35, 593-596.
Glaeser, E.L. and Berry, C. R. (2006). Why are smart places getting smarter?, Taubman Cente Policy Brief
2006-2, Cambridge MA: Taubman Centre.
Graham, S. and Marvin, S. (1996). Telecommunications and the city: electronic spaces, urban place, London:
Routledge.
Hollands, R. G.(2008). Will the real smart city please stand up?, City, 12 (3), 303- 320.
Komninos. N. (2002). Intelligent cities: innovation, knowledge systems and digital spaces. London: Spon Press.
Lucas, R. E. (1988), On the mechanics of economic development, Journal of Monetary Economics, 22, 3-42.

Mahizhnan, A. (1999). Smart cities: the Singapore case, Cities, 16(1), 13-18.
Mankiw, N. G., Romer, D. and Weil, D. N. (1992). A contribution to the empirics of economic growth, The
Quarterly Journal of Economics, 107 (2), 407-37.

Mayer, H. (2007). What is the role of the university in creating a high-technology region?, Journal of Urban
Technology, 14 (3), 33-58.

Nijkamp, P. (2008). E pluribus unum, Research Memorandum, Faculty of Economics, VU University

Amsterdam.
OECD EUROSTAT (2005). Oslo manual, Paris: Organization for Economic Cooperation and Development
Statistical Office of the European Communities.
Poelhekke, S. (2006). Do Amenities and Diversity Encourage City Growth? A Link Through Skilled Labor,
Economics Working Papers ECO2006/10, San Domenico di Fiesole, Italy: European University Institute.
Redding, S.J. and Sturm, D.M. (2008). The costs of remoteness: evidence from German division and
reunification, The American Economic Review, 98 (5), 1766-1797.
Shapiro, J. M. (2008). Smart cities: quality of life, productivity, and the growth effects of human capital, The
Review of Economics and Statistics, 88 (2), 324-335.
Southampton City Council (2006). Southampton on-line,
http://www.southampton.gov.uk/thecouncil/thecouncil/you-and-council/smartcities/ (accessed on 13 March
2009).
The Science Museum (2004). Urban development,
http://www.makingthemodernworld.org.uk/learning_modules/geography/04.TU.01/?section=2 (accessed on
3 April 2009).