176 World Review of Science, Technology and Sust. Development, Vol. 11, No.

2, 2014

A new approach to measuring the liveability of cities:

the Global Liveable Cities Index

Tan Khee Giap*

Asia Competitiveness Institute,
Lee Kuan Yew School of Public Policy,
National University of Singapore,
469A Bukit Timah Road,
Level 2 Wing A, 259770, Singapore
E-mail: spptkg@nus.edu.sg
*Corresponding author

Woo Wing Thye

Economics Department,
University of California,
One Shields Avenue,
Davis, California 956161, USA
E-mail: sppwwt@nus.edu.sg

Grace Aw
Mathematics and Statistics Department,
Curtin University,
Kent Street, Bentley 6102, Australia
E-mail: sppgael@nus.edu.sg

Abstract: This paper advocates a new measure of liveability – the Global

Liveable Cities Index (GLCI) – to rank the world’s major cities. The GLCI
advances the measurement of the ‘liveability’ construct by taking into account
the multidimensional sensibility of diverse groups of ordinary persons across
64 cities. The paper also compares the GLCI with six well known city indices
from the literature and finds that the GLCI performs as a relevant liveability
indicator for diverse groups of ordinary persons. In addition, the paper conducts
policy simulations to find that city planners interested in attracting talent of a
given personality type can invest in areas with low scores in the GLCI.

Keywords: economic competitiveness; liveability; city benchmarking.

Reference to this paper should be made as follows: Giap, T.K., Thye, W.W.
and Aw, G. (2014) ‘A new approach to measuring the liveability of cities: the
Global Liveable Cities Index’, World Review of Science, Technology and
Sustainable Development, Vol. 11, No. 2, pp.176–196.

Biographical notes: Tan Khee Giap is co-Director in Asia Competitiveness

Institute and Associate Professor in Lee Kuan Yew School of Public Policy,
National University of Singapore, Singapore.

Copyright © 2014 Inderscience Enterprises Ltd.

 A new approach to measuring the liveability of cities 177

Woo Wing Thye is Professor in Department of Economics, University of

California, Davis and Adjunct Senior Research Fellow at ACI-LKYSPP,
Singapore.

Grace Aw is Senior Lecturer in Department of Mathematics and Statistics,

Curtin University of Technology, Australia and Adjunct Research Fellow at
ACI-LKYSPP, Singapore.

1 Introduction

The last few decades have seen the world becoming more urbanised. With more than half
of the world’s population living in urban areas in 2011 from about 30% in 1950, it is
projected that the urban areas in the world will host 70% of the world’s population by
2050. While considerable heterogeneity exists in the levels of urbanisation across
different regions of the world, the Asian region stands out as having the potential to see
half of its population live in the urban areas by 2020 (United Nations, 2011).
Furthermore, the future urban population is expected to be increasingly concentrated in
large cities of one million or more inhabitants, with the megacities consisting of at
least 10 million inhabitants experiencing the largest percentage increase. As the
United Nations (2011) notes, such increasing urban concentration in very large cities is a
relatively new phenomenon that the world is experiencing which also highlights the
growing need to focus on the quality of living in such cities.
A central feature of this rising urbanisation has been the growing importance of cities
as a locus of economic activity. Over the last few decades, with globalisation leading to
the intensification of cross-border mobility not just of goods and financial capital but also
of labour and human talents, there has also been a marked rise in interconnectivity
among cities, primarily due to advancements in transportation and telecommunication
technology. To be sure, a combination of attractive infrastructure and supporting
regulatory environments have helped cities compete for global talents, resources and
capital, positioning themselves as a platform for innovations to occur and ideas to grow.
‘Liveability’ is one key characteristic of cities that enable them to attract a
disproportionate amount of the globally-mobile resources (such as talents, high net worth
individuals, investors, innovators, entrepreneurs, and capital) that are recognised to make
positive contributions to economic growth, economic resilience, global political
influence, world agenda-setting power, socio-cultural innovation, and international
lifestyle impact. As competition among cities is considered as strategic as competition
between nations, the city that fares well in the competition becomes an epicentre for
advanced economic and cultural activities while those that lack the required standards of
competitiveness shrink economically and become irrelevant (Tan et al., 2012).
Thus, the growing popular interest in the liveability of cities [examples of bestsellers
on cities in the last two decades are Sassen (1991) and Glaeser (2011)] has also led to an
increasing desire to rank the liveability of cities that will help policy makers frame
appropriate policies. The accelerated globalisation has also reinforced our understanding
that the agglomeration of activities by cities constitutes powerful growth engines. In the
words of Glaeser (2011) ‘cities magnify humanity’s strength’ as they improve enable
socio-economic mobility by creating opportunities, and inducing innovation by easing
178 T.K. Giap et al.

face-to-face engagements. An increasing recognition of the importance of global

competition amongst cities in attracting all forms of capital has also been acknowledged
by the OECD (2007) in its Competitive Cities Report. The rise in global competition is
also reflected in the increasing popularity of city benchmarking as well, of which notions
of liveability form an important component.
In recent years, a number of liveability and benchmarking indices and studies have
been published to assess the relative position of various global cities against each other in
various categories. These liveability measures are typically used as a tool to make
comparisons between cities with various outcome scores receiving widespread media
attention. However, none of the existing indices take the perspective of the ordinary man
living in that city and model this ordinary person as having multi-dimensional
sensibilities towards issues like economic well-being, social mobility, personal security,
political governance, environmental sustainability, and aesthetics. This paper advocates a
new measure of liveability to rank the world’s major cities – the Global Liveable Cities
Index (GLCI) first introduced by Tan et al. (2012). Section 2 begins with a discussion
about the various definitions of liveability and briefly examines the limited literature.
Section 3 provides the theoretical basis for the proposed GLCI. Section 4 performs a
comparison of the proposed index with six well-known and widely employed indicators
from the literature and Section 5 concludes.

2 Defining ‘liveability’ – related literature

There is no established theoretical framework laying out a uniform definition of

liveability. As such, the literature has a handful of empirical studies that broadly involve
a direct comparison of a composite measure of liveability over different geographic areas
and generate rankings categorising the liveability of cities.
However, a broad understanding of the notion of liveability appears to have emerged,
though there is no consensus on the specificities of its scope. A very useful working
definition of urban liveability comes from Vuchic (1999, p.7) who defines it as
“…generally understood to encompass those elements of home, neighborhood, and
metropolitan area that contribute to safety, economic opportunities and welfare, health,
convenience, mobility, and recreation”. The concept of livability is clearly more a
qualitative construct representing a set of characteristics that relate to the attractiveness of
an area as a ‘desirable’ place to live, work, invest, and conduct business.
Liveability in many ways could be viewed as encompassing a wide range of issues
relating to overall ‘quality of life and well-being’. It is intuitive to understand it as a
‘place-based’ concept that generally refers to those elements of a home, neighbourhood,
or city that contribute to quality of life and well-being. Quality of life and well-being are
closely related concepts that relate to the dimensions on which an individual’s living
condition or state can be measured. They can range from the more objective indicators of
economic wellbeing, such as human capital, to the more subjective quality of life
indicators that include social capital, qualitative expressions of personal satisfaction and
the like (Ley and Newton, 2010).
Interestingly, liveability as represented by human well-being and the quality of city’s
physical environment is derived in turn from the performance of key urban systems and
processes in the cities where people live and work. This is what that gives rise to several
indicators in the literature that are intended operate as broad proxies to measure and judge
 A new approach to measuring the liveability of cities 179

the degree of liveability of a city/area. Indicators as to whether the housing market

provides appropriate and affordable housing or whether the transport system provides a
high level of mobility and connectivity without high levels of car dependency etc are
indicators that help measure the quality of life in a city are examples of such proxies
measuring liveability. Thus to put it in another way, the notion of ‘liveability’ is used to
describe the overall contribution of the urban environment in influencing the quality of
life or wellbeing of residents (Urbis, 2008).
That being said, liveability as used in the literature has also been broadly equated
with sustainable development, for most indicators that are used to measure liveability
invariably fall under the ambit of a ‘cleaner, safer and greener” environment that falls
under the rubric of sustainable development in the policy discourse. In fact, as Ley and
Newton (2010) emphasise, the concept of liveability is considered one of the four key
domains central to the notion of sustainable urban development, the others being
environmental sustainability, economic performance, and good governance. As some
studies show, in many advanced economies like the UK for instance, liveability has been
adopted in a much narrower and more operational sense and has to do with high degrees
of ‘cleanliness, safety and green-friendliness’. Though such a broad definition may well
be considered an ‘umbrella’ term that encompasses a number of inter-related issues, its
overwhelming focus appears to be on the local environment.
However, as Woolcock (2009) argues, the link between liveability and sustainable
development is not very clear either. In some cases the two terms are being used
interchangeably while in other contexts, liveability is being considered as a subset of a
sustainable city. A literature scan by Lyndhurst (2004) for example on the various
definitions of liveability, concludes that there is a general lack of consensus in the
research and policy literature about the possible interplay between liveability and
sustainable development and that little attention is being paid to the possibility that both
the concepts could have ‘mutually reinforcing’ as well as ‘potentially conflicting’
elements.

3 Analytical basis of the GLCI

City benchmarking can be conceptualised as measuring and monitoring the performance

of cities against a number of comparable and/or best practice cities. As Luque-Martinez
and Munoz-Leiva (2005) summarise, city benchmarking is “the systematic continuous
method that consists of identifying, learning and implementing the most effective
practices and capacities from other cities in order to improve one’s own city to improve
its action in what it offer”.
While a considerable number of indices on city benchmarks already exist in the
literature, none of them model liveability from the perspective of an ordinary person. We
briefly review the following five empirical frameworks discussed in detail in Tan et al.
(2012), which are most relevant to developing the empirical framework for GLCI:
• World Competitiveness Yearbook
• Mercer Human Resource’s Worldwide Quality of Living Survey
• Yale & Columbia University’s 2005 Environment Sustainability Index
180 T.K. Giap et al.

• The Global City Indicators Program

• Indicators of Sustainable Development: Guidelines & Methodology
The World Competitiveness Yearbook’s focus is on countries and their economic
competitiveness, while the Mercer study’s focus is on the quality of living across cities
for expatriates. As for the Yale & Columbia study, its focus is on environment
sustainability with an emphasis on ‘green’ indicators and international initiatives but does
not address the issue of quality of life.
The Global City Indicators Program aims to provide an established set of city
indicators with a globally standardised methodology that allows for comparability of
performance. The primary focus is to measure the social well-being of citizens and its
indicators range from healthcare and transportation to technology. However, it has
limited coverage of government leadership and factors like policy making and its
implementation as well as corruption and foreign affairs are all neglected.
Indicators of Sustainable Development are an attempt to measure the progress
towards sustainable development and encompass themes such as poverty, governance,
health, education and economic development. Though this index provides a multi-
dimensional nature for measuring sustainability, its approach used is more suited for
measuring a single nation’s development rather than for making comparisons between
cities.
Since human nature is complex, the concept of liveability is necessarily a complex
one. At the very least, the concept of ‘liveability’ has to be multi-dimensional in the same
way that human nature is. In developing the GLCI, we model an ordinary person with
multi-dimensional sensibilities towards issues like economic well-being, social mobility,
personal security, political governance, environmental sustainability, and aesthetics for a
more representative coverage of major cities around the world. The GLCI offers one such
index which could be highly useful for policy making.
The ordinary-resident’s-perspective approach of GLCI makes GLCI substantially
different from most other well-known liveability rankings of cities like the Mercer
Quality of Living Survey, and the Knight Frank Global Cities Survey. The GLCI
explicitly takes into account a comprehensive list of the everyday concerns of the
ordinary household: the maintenance of law and order, the availability of affordable
healthcare, the average quality of the public school system, the accessibility to tertiary
level training, and the adequacy of the mass transit infrastructure. In short, this paper is
an attempt to broaden the scope of earlier studies by others within an encompassing
theoretical as well as empirical framework, laid out in Tan et al. (2012).
Tan et al. (2012) ranks the liveability of cities by capturing the multi-dimensional
character of liveability in five themes to operationalise measurement. These five themes
have their theoretical basis in the social sciences, humanities and natural philosophy.
Specifically, they are:
1 economic vibrancy and competitiveness
2 domestic security and stability
3 socio-cultural conditions
4 public governance
5 environmental friendliness and sustainability.
 A new approach to measuring the liveability of cities 181

The sequence of the five themes is not in any order of perceived priority or indicative of
their relative importance. Together, they provide a conceptual framework of liveability in
accordance with various depictions of the nature of man in the social sciences and
humanities (with all the five categories equally weighted in the index).
Given the fundamental importance of ecological sustainability in the concept of
liveability, we present a discussion of an important issue that differentiates our
worldview from that of many other studies that rank cities or countries. The defining
issue that divides us from many others is about how to measure ecological sustainability
at the city level and at the country-level. Indicators for category 1 are the usual hard
economic data related to its openness and pro-business policies which should be readily
available in the public domain. Indicators in category 2 would typical involve proxies
such as crime rate, social harmony, civil unrest, threats to domestic security and stability.
Indicators for category 3 on quality of life and diversity would entail public services such
as affordable health care, education, public housing, sanitation and transportation as well
as income disparity, demography burden and community cohesion. Indicators on
category 4, being more difficult to quantify, would rely heavily on survey data pertaining
to quality of government, policy effectiveness, transparency and accountability, fair and
efficient justice system.
Indicators in category 5 would involve technical indicators usually covering
pollution, green spaces, recycling rate and water quality (see Tan et al., 2012). While the
first four indicators are generally well-discussed in the literature, we focus on category 5.

3.1 Environmental friendliness and sustainability

We take the viewpoint that sustainability at the city level should be measured by the
extent that a city implements the principle of ‘think globally and act locally’.
Specifically, our analytical position is that, in an interdependent world of nation states,
there are two components that should be accounted for while defining ecological
sustainability:
a ecological sustainability always means ecological sustainability at the global level
b ecological sustainability at the local level should not always be equated with local
self-sufficiency in meeting the needs of the local community.
The first component tells us to ‘think globally’. The second component tells us ‘to act
locally in a way that is consistent with maximum global welfare’ because if the local
action is not consistent with the global optimum, then the whole world is made worse off.
To see the saliency of the second component, consider the recent study conducted by
the Fisheries Centre at the University of British Columbia (UBC, 2012) which computed
an ‘Eco2 Index’ to aggregate the economic deficit and the ecological deficit of a country.
The ecological deficit measures “resource consumption and waste produced by a country
in comparison to its carrying capacity as expressed in locally available resources such as
agricultural land and energy”. The primary problem with this UBC index is that a country
like the UK that imports most of its food and energy would thus have a bad score on its
ecological sustainability performance, while a food exporter like Myanmar would have a
high score on ecological sustainability. Physically small countries (and city states in
particular), like Bermuda, Curacao, Malta, Singapore and Switzerland, that have gotten
rich through active participation in international trade are hence by definition
182 T.K. Giap et al.

ecologically un-sustainable. Herein lies the structural flaw of the UBC index: the Eco2
Index holds that self-sufficiency is a desired condition in and of itself. Such a viewpoint
is fundamentally a reflection of a survivalist creed: today’s friend can be tomorrow’s
enemy and hence a city should never be reliant on the possible fickleness of others. In
short, the Eco2 Index would identify the best world economic order to be an autarkic
world order. Furthermore, since a city normally relies on the countryside for food, the
application of the Eco2 Index at the city-level would yield the conclusion that the best
national economic system would be a nation of self-subsistence farmers.
We explicitly reject the survivalist philosophy that the potential for self-sufficiency in
food and energy production is the appropriate measure of the sustainability of a city or
country. This survivalist mentality is at odds with the insights of Adam Smith and
David Ricardo that identify specialisation in production as the basis for wealth creation in
normal times. It is only during abnormal times (like the periods of worldwide conflicts)
where there is virtual suspension of international trade is the survivalist criteria the
correct indicator for sustainability. But we are not interested in doing a ranking of
liveability that applies only during abnormal times when autarky is externally imposed.
The kind of country/city ranking studies that is exemplified by the report of the
Fisheries Centre of the UBC (2012) should more appropriately be called ‘rankings of
survival-bility’ or, simply, ‘rankings of self-sufficiency’. In our thinking, however, living
life to the fullest is meaningful only if life is more than mere survival.
The survivalist interpretation of sustainability is really unsuitable to the modern
world. If drastic climate changes were to occur abruptly in the food-exporting parts of the
world and trigger protectionism, a nation of subsistence farmers would, indeed, escape
largely unscathed from the meltdown of the world food market. However, this autarkic
nation would still not survive a global nuclear Armageddon. The present reality is that the
practice of self-sufficiency will not guarantee survival in a world with nuclear arms. So if
we update the survivalist philosophy to present circumstances, the Eco2 Index should
measure sustainability of a country by, one, the amount of effort that country’s
government puts into the campaign for a global ban on nuclear weapons; and, two, by
how far the country is from the closest country with nuclear arms (because this closest
nuclear-armed country has a higher probability of being a first-strike target by other
nuclear-armed countries).
We would like to propose in this paper that ecological sustainability is better
guaranteed by the practice of ‘think globally, and act locally’. We make this point by
considering the case of the emission of greenhouse gases like CO2 by a country. It is clear
that the best contribution that a country could make to the global situation (from thinking
globally) is to minimise its annual emission of GHG, which we denote as G, measured in
parts per million (ppm) per year.
Now, what is to be done locally in order to reduce G (or, at least, keep the growth rate
of G at a minimum)? The value of G, the additional amount of GHG in the air each year,
is determined by the PIES-in-the-sky equation:

G = P * I * E *S

where

P population size

I income per capita, i.e., GDP/P

 A new approach to measuring the liveability of cities 183

E energy inefficiency defined as amount of energy (in Joules, J) consumed in

producing a unit of GDP, i.e., J/GDP
S soiling capacity of the energy used defined as amount of GHG added by each unit of
energy consumed, i.e., G/J
So we have:
G = P * (GDP / P) * (J / GDP) *(G / J)

or
G = P * (GDP / P) * (G / GDP).

For any emerging economy, responsible global citizenship would have its policy makers
enact policies that would lower (J/GDP), (G/J) and P, i.e., increase energy efficiency,
switch to green energy, and strengthen the family planning program. As the average
income (I) of an emerging economy is still way below the levels in Western Europe and
North America, the obligation of the government to its own citizens is to continue to let
output (I) grow as fast as conditions permit. The faster that output grows, the more the
government should do to reduce energy inefficiency (E), the use of dirty energy (S) and
the rate of population growth (P). In short, the policy target that flows naturally from
responsible global citizenship (‘thinking globally and acting locally’) and from the right
of countries (especially of the poorer countries) to grow is the emission-GDP ratio
(G/GDP) rather than the per capita amount of emission (G/P).1
The policy agenda that follows from the survivalist interpretation of sustainability
differs significantly from our above policy agenda. It seems straightforward that the
implementation of self-sufficiency in food and energy would reduce the growth of GHG
emission by reducing the growth of output (I) because output is creased by international
trade. However, this GHG-reduction outcome from switching to a self-sufficiency regime
is far from certain. In the cases of China and India, energy sufficiency would require
these two countries to switch from imported oil to domestic coal, and to generate more
hydropower by building more Three Gorges Dam-type of projects.2 This means that the
only way that India and China could attain energy sufficiency without emitting even
more GHG and tearing up more of their natural environments would be if there were
revolutionary technological breakthroughs in solar power, wind power, and
carbon-capture-and-sequestration. Such technological breakthroughs are, however, just as
likely to occur under the ‘think globally, and act locally’ policy regime as under the
self-sufficiency policy regime. Of course, the actual measure of ‘environmental
friendliness and sustainability’ (‘the state of the natural environment and its
management’) used in constructing our GLCI takes many more factors into consideration
and not just the value of (GHG/GDP). The above discussion on GHG is only meant to
illustrate the basic differences in philosophy that guide the measurement of ecological
sustainability in our GLCI study and some other major studies, e.g. the global-citizen
approach versus the survivalist approach. Our measure of ‘environmental friendliness and
sustainability’ is constructed from 15 environmental sustainability indicators at the city
level; and these 15 indicators could be grouped under three categories:
1 extent of air and water pollution
2 extent of depletion of natural resources
184 T.K. Giap et al.

3 extent of government involvement in efforts to protect the environment.

3.2 Indicators of GLCI

The indicators used to reflect each of the five categories in the GLCI are shown in
Tables 1–5 respectively.
Table 1 Indicators for economic vibrancy and competitiveness

Economic performance Economic openness Infrastructure

GDP Foreign direct investment Telephone lines
(percentage of GDP) (subscribers per 100 people)
Real GDP growth rate Trade to GDP ratio Computer ownership
(per 1,000 people)
Labour productivity per hour State ownership of Level of internet access
enterprises (index) (percentage population)
Household consumption Prevalence of trade
expenditure per capita barriers (index)
Unemployment rate Number of free trade
agreements
Resilience of economy (index) Ease of doing business (index)
Gross fixed capital formation Prevalence of foreign
(percentage of GDP) ownership (index)
Growth rate of consumer price Tourism receipts
index (percentage of GDP)
Debt to gross national Economic freedom (index)
income ratio
Hotel occupancy rates
International tourist arrivals

Table 2 Indicators for environmental friendliness and sustainability

Pollution Depletion of natural resources Environment initiatives

Greenhouse gas emissions Electricity generated from Participation in selected
renewable sources (percentage international environmental
of total electricity generated) agreements (out of total of 11)
Sulphur dioxide emission Consumption of oil per day Stringency of environmental
regulations (index)
CO2 emissions in 2006 Threatened species (percentage Terrestrial protected area
of total animal species) (percentage of total land area)
Quality of the natural Protected marine
environment (index) area (percentage of
total marine area)
Water pollution Enforcement of
(kilograms per day per worker) environmental
regulation (index)
Nitrogen oxide emission
Particulate matter
concentration
 A new approach to measuring the liveability of cities 185

Table 3 Indicators for domestic security and stability

Crime rate Threats to national stability Civil unrest

Number of homicide cases Business costs of Severity of political
(per 10,000 capita) terrorism (survey) violence (index)
Number of new drug offences Fatalities of terrorist attacks Conflicts of ethnic, religious,
(per 100,000 capita) (per million capita) regional nature (index)
Business cost of crime and Natural disaster death toll Violent social
violence (index) (per million capita) conflicts (index)
Reliability of police
services (index)

Table 4 Indicators for socio-cultural conditions

Income quality Diversity and

Medical and Housing, sanitation
Education and demographic community
healthcare and transportation
burden cohesion
Infant mortality Quality of Percentage of GINI index Percentage of
rate education urban population foreigners/
system (index) living in slums percentage of
immigrants
Life expectancy Tertiary Percentage of Number of hours Number of
enrolment rate population using worked per year religions (index)
improved sanitation (index)
Government Government Population using Human poverty Attitudes
health expenditure expenditure on an improved index towards foreign
per capita education water source visitors (index)
Number of Higher Quality of ground Child dependency
hospital beds (per education transport network ratio
1,000 population) achievement (index)
Density of Quality of roads Old age
physicians (per (index) dependency ratio
10,000 population)
Quality of railroad
infrastructure (index)
Quality of electricity
supply (index)

Table 5 Indicators for political governance

Policy making and Transparency and

Government system Corruption
implementation accountability
Government effectiveness Functioning of Transparency of Control of
(index) government system economic policy corruption
(index) (index) (index)
Government consumption Effectiveness of judicial Voice and Corruption
expenditure (percentage system (index) accountability perceptions index
of GDP) (index) (index)
Collected total tax revenues Quality of e-government
(percentage of GDP) (index)
Regulatory quality (index) Political stability no
violence (index)
Rule of law (index)
186 T.K. Giap et al.

4 City ranking and policy simulations

Table 6 shows the overall ranking for 64 global cities [with details for individual ranking
by the five categories found in Tan et al. (2012) Chapter 3]. Of the 64 global cities, five
Asian cities are in the top 20. They are Singapore, Hong Kong, Osaka-Kobe, Tokyo and
Yokohama.
Table 6 Overall ranking for 64 global cities

Overall Overall
City Region liveability City Region liveability
Score Rank Score rank
Geneva Europe 3.40 1 Kuala Lumpur Asean 32.00 32
Zurich Europe 4.60 2 Rome Europe 34.00 34
Singapore Asean 5.60 3 Amman Mid East 36.60 35
Copenhagen Europe 7.00 4 Jerusalem Asia 37.00 36
Helsinki Europe 7.00 4 Sao Paulo S. America 43.40 37
Luxembourg Europe 7.80 6 Riyadh Mid East 44.00 38
Stockholm Europe 8.20 7 Shanghai Asia 45.00 39
Berlin Europe 11.20 8 Nanjing Asia 45.20 40
Hong Kong Asia 11.20 8 Bangkok Asean 45.80 41
Auckland Oceania 11.60 10 Shenzhen Asia 45.80 41
Melbourne Oceania 11.60 10 Ahmedabad Asia 46.00 43
Sydney Oceania 12.00 12 Cairo Mid East 46.00 43
Paris Europe 12.40 13 Tianjin Asia 47.40 45
Vancouver N. America 16.20 14 Beijing Asia 47.80 46
Amsterdam Europe 16.80 15 Chennai Asia 48.20 47
Osaka-Kobe Asia 17.80 16 Guangzhou Asia 48.20 47
New York N. America 18.20 17 Pune Asia 48.20 47
Tokyo Asia 18.60 18 Mexico City N. America 48.40 50
Los Angeles N. America 18.80 19 Damascus Mid East 48.60 51
Philadelphia N. America 21.40 20 Chongqing Asia 48.80 52
Yokohama Asia 21.40 20 Hanoi Asean 48.80 52
Boston N. America 21.60 22 Ho Chi Minh City Asean 48.80 52
London Europe 21.60 22 Bangalore Asia 49.00 55
Chicago N. America 22.40 24 Mumbai Asia 49.00 55
Washington DC N. America 22.80 25 Delhi Asia 50.20 57
Barcelona Europe 23.20 26 Buenos Aires S. America 50.60 58
Taipei Asia 24.00 27 Istanbul Mid East 52.20 59
Prague Europe 25.80 28 Karachi Mid East 53.00 60
Seoul Asia 26.20 29 Phnom Penh Asean 53.80 61
Madrid Europe 27.00 30 Moscow Europe 55.20 62
Incheon Asia 27.40 31 Manila Asean 56.60 63
Abu Dhabi Mid East 32.00 32 Jakarta Asean 57.40 64
Source: Tan et al. (2012, Table 21, pp.52–53)
 A new approach to measuring the liveability of cities 187

In particular for economic vibrancy and competitiveness, Hong Kong and Singapore are
ranked fourth and fifth, respectively, among the top 20 global cities. Of the other
categories, Singapore tops the list only for domestic security and stability, while being
third in political governance, fifth in social-cultural conditions and 14th in environmental
friendliness and sustainability.
Table 7 Overall ranking for 64 global cities after simulation

Overall Overall
City Region liveability City Region liveability
Before After Before After
Geneva Europe 1 1 Kuala Lumpur Asean 32 23
Zurich Europe 2 1 Rome Europe 34 29
Singapore Asean 3 1 Amman Mid East 35 30
Copenhagen Europe 4 1 Jerusalem Asia 36 26
Helsinki Europe 4 1 Sao Paulo S. America 37 32
Luxembourg Europe 6 4 Riyadh Mid East 38 35
Stockholm Europe 7 1 Shanghai Asia 39 34
Berlin Europe 8 2 Nanjing Asia 40 34
Hong Kong Asia 8 4 Bangkok Asean 41 36
Auckland Oceania 10 2 Shenzhen Asia 41 35
Melbourne Oceania 10 5 Ahmedabad Asia 43 37
Sydney Oceania 12 4 Cairo Mid East 43 35
Paris Europe 13 8 Tianjin Asia 45 37
Vancouver N. America 14 6 Beijing Asia 46 37
Amsterdam Europe 15 8 Chennai Asia 47 37
Osaka-Kobe Asia 16 9 Guangzhou Asia 47 37
New York N. America 17 12 Pune Asia 47 37
Tokyo Asia 18 8 Mexico City N. America 50 37
Los Angeles N. America 19 12 Damascus Mid East 51 37
Philadelphia N. America 20 14 Chongqing Asia 52 37
Yokohama Asia 20 14 Hanoi Asean 52 35
Boston N. America 22 14 Ho Chi Minh City Asean 52 37
London Europe 22 10 Bangalore Asia 55 37
Chicago N. America 24 14 Mumbai Asia 55 37
Washington DC N. America 25 14 Delhi Asia 57 37
Barcelona Europe 26 15 Buenos Aires S. America 58 37
Taipei Asia 27 20 Istanbul Mid East 59 37
Prague Europe 28 16 Karachi Mid East 60 37
Seoul Asia 29 16 Phnom Penh Asean 61 37
Madrid Europe 30 15 Moscow Europe 62 53
Incheon Asia 31 19 Manila Asean 63 39
Abu Dhabi Mid East 32 22 Jakarta Asean 64 41

While rankings are useful, a notable innovation of the Tan et al. (2012) study lies in
policy simulations to explore the extent to which the city is able to improve on its
188 T.K. Giap et al.

liveability ranking. The simulation is based on two policy assumptions, one being that
each city will work on areas where their rankings are worst, in order to improve. This is
identified by their lowest (worst-performing) 20% of the indicators which are selected
from the entire list of indicators regardless of which category they belong to.3 The bottom
20% is picked to focus on because of limited resources to work on all areas concurrently.
The second policy assumption is that, after identifying the 20% most lagging
indicators, policies as devised for implementation to raise their scores are simulated to the
‘average’ score of a particular indicator for all cities (computed using the original data).
As the simulation is static where the ranks are computed assuming only a particular city
improves at a time holding all other cities’ scores constant, all cities’ rankings after the
simulation will never decline.
The simulation shows the potential for improvement by a change in overall rankings
before and after simulation. Apart from their usefulness to policy-makers, results of the
simulation will also help the business community, potential investors and residents.
Information on potential liveability will enable informed choices and decision-making on
business ventures, including migration for better competitiveness by all.
Table 7 presents the results of policy simulations undertaken. As we can see from the
table, Singapore emerges as the only Asian city tying as first with Geneva, Zurich,
Copenhagen, Helsinki and Stockholm in the overall rankings. Equally interesting is the
dramatic improvement or catch-up by many cities like Chicago from 34 to 5, Shanghai
from 45 to 20 and Amman from 48 to 20. The rise of Abu Dhabi from 32 to 22 is another
proof of how poor performance before simulation is greatly enhanced by policies.

4.1 GLCI and other city ranking indices – a comparison

In this section, we present a comparison of the GLCI with six other popular city indices
used in the literature , by picking the top 10 cities of each index. The other six indices
considered for the study are:
1 the Global Power City Index by the Mori Memorial Foundation (2011), i.e., the
Mori Index
2 the Global Cities Index jointly by the Foreign Policy magazine, The Chicago Council
on Global Affairs, A.T. Kearney (2010), i.e., the Foreign Policy Index
3 the Global Cities Index jointly by Frank (2011a) and (2011b) and Citi Private Bank,
i.e., the Knight Frank Index
4 the Global City Competitiveness Index jointly by the Economist Intelligence Unit
(EIU, 2011a) and CitiGroup, i.e., the EIU-Competitiveness Index
5 the Quality of Living Index by Mercer (2011), i.e., the Mercer Index
6 the Liveability Ranking issued by EIU (2011b), i.e., the EIU-Liveability Index.
Table 8 reports the ten top-rated cities according to the above six city indices and GLCI.4
Given that there is some overlap in the contents of each index, the seven city indices
contain 34 different cities. The number of times that a city appears in the indices, i.e., a
city citation record is shown in Table 9 while Table 10 shows how many indices that a
particular index overlaps with, i.e., an index citation record.
 Table 8

Source of index Mori Memorial Foreign Policy and Knight Frank and Economist Economist Mercer Centre for Liveable
Foundation AT Kearney Citibank Intelligence Unit Intelligence Unit Cities
and Citibank
Name of index Global Power City Global Cities Global Cities Global City Liveability Quality of Life Global Liveable
Competitiveness Cities
Sample size 35 65 40 120 140 221 64
Year 2011 2010 2011 2011 2011 2011 2010
Rank (a) (b) (c) (d) (e) (f) (g)
1 New York New York New York New York Melbourne Vienna Geneva
2 London London London London Vienna Zurich Zurich
3 Paris Tokyo Paris Singapore Vancouver Auckland Singapore
The top 10-ranked cities in each city index

4 Tokyo Paris Tokyo Paris Toronto Munich Copenhagen

5 Singapore Hong Kong Brussels Hong Kong Calgary Dusseldorf Helsinki
6 Berlin Chicago Los Angeles Tokyo Sydney Vancouver Luxembourg
7 Seoul Los Angeles Singapore Zurich Helsinki Frankfurt Stockholm
A new approach to measuring the liveability of cities

8 Hong Kong Singapore Beijing Washington DC Perth Geneva Berlin

9 Amsterdam Sydney Toronto Chicago Adelaide Bern Hong Kong
10 Frankfurt Seoul Berlin Boston Auckland Copenhagen Auckland
10 Melbourne
189
190 T.K. Giap et al.

Table 9 Citation record of each city

City Frequency mentioned Listed in top ten in the city index of

Singapore 5 Mori, Foreign Policy, Knight Frank, EIU-C, GLCI
Hong Kong 4 Mori, Foreign Policy, EIU-C, GLCI
London 4 Mori, Foreign Policy, Knight Frank, EIU-C
New York City 4 Mori, Foreign Policy, Knight Frank, EIU-C
Paris 4 Mori, Foreign Policy, Knight Frank, EIU-C
Tokyo 4 Mori, Foreign Policy, Knight Frank, EIU-C
Auckland 3 EIU-L, Mercer, GLCI
Berlin 3 Mori, Knight Frank, GLCI
Zurich 3 EIU-C, Mercer, GLCI
Chicago 2 Foreign Policy, EIU-C
Copenhagen 2 Mercer, GLCI
Frankfurt 2 Mori, Mercer
Geneva 2 Mercer-GLCI
Helsinki 2 EIU-L, GLCI
Los Angeles 2 Foreign Policy, Knight Frank
Melbourne 2 EIU-L, GLCI
Seoul 2 Mori, Foreign Policy
Sydney 2 Foreign Policy, EIU-L
Toronto 2 Knight Frank, EIU-L
Vancouver 2 EIU-L, Mercer
Vienna 2 EIU-L, Mercer
Adelaide 1 EIU-L
Amsterdam 1 Mori
Beijing 1 Knight Frank
Bern 1 Mercer
Boston 1 EIU-C
Brussels 1 Knight-Frank
Calgary 1 EIU-L
Dusseldorf 1 Mercer
Luxembourg 1 GLCI
Munich 1 Mercer
Perth 1 EIU-L
Stockholm 1 GLCI
Washington DC 1 EIU-C
Notes: EIU-C = EIU-Competiveness
EIU-L = EIU-Liveability
 A new approach to measuring the liveability of cities 191

Table 10 Citation record of each city index

Number of indices with which this Number of cities in this index

City index
index has overlap which appears in other indices
GLCI 6 9
Foreign Policy 5 9
Mori 5 8
Knight Frank 5 8
EIU-Competitiveness 5 7
EIU-Liveability 4 6
Mercer 3 5

Together, Tables 8, 9 and 10 allow two interesting conclusions. First, there is not too
much overlap in top-10 cities of seven indices. Singapore is the most cited city appearing
in five of the seven indices, and Hong Kong, London, New York City, Paris and Tokyo
appear in four indices. Twenty-eight out of 34 cities appear in less than half of the
indices.
Second, seven indices fall into three categories: clout club indices, comfort club
indices and balanced club indices.
• The clout club: There is a very large overlap of the top-10 cities in the Mori Index,
Foreign Policy Index and Knight Frank Index. Their top four cities are the same,
namely, New York, London, Paris and Tokyo. These four cities are also in top-10
lists of the EIU-Competitiveness Index, but not in the EIU-Liveability Index, the
Mercer Index and the GLCI. Since New York, London, Paris and Tokyo are strong
global centres of financial power, economic prowess, political influence, social status
and cultural leadership, we can describe the Mori Index, Foreign Policy Index,
Knight Frank Index and EIU-Competitiveness Index as ‘clout club indices’ and their
top-10 cities as clout club cities.
• The comfort club: Both the EIU-Liveability and the Mercer indices originate as
guidelines to the personnel departments in multinational corporations in the
dispensation of hardship pay. As these two indices emphasise ‘pleasant living’, we
call them ‘comfort club indices’; and the top 10 cities in each index as the comfort
club cities. We note, however, that these two self-professed ‘pleasantness-of-living’
indices have very little overlap. Their respective top-10 lists share only Vienna in
common. The ranking of the comfort club indices reflects clearly the fact that what is
pleasant depends on one’s taste, as evidenced by the differences in the common
language trait in the respective top 10 cities identified by the Mercer Index and the
EIU-Liveability. Six of the top 10 cities in the Mercer Index (Vienna, Zurich,
Munich, Dusseldorf, Frankfurt, and Bern) are German-speaking, and two of them
(Copenhagen and Geneva) are right on the border both of the Germanic cultural
sphere and the Germanic geographical sphere.
• The balanced club: The GLCI overlaps with the clout club in four cities: Singapore
(4), Hong Kong (3), Berlin (2) and Zurich (1). The number in the parenthesis reports
the number of times that this GLCI top-10 city appears in the clout club’s top-10
lists. This degree of overlap between GLCI and the clout club indices is substantially
higher than the overlap between the clout club indices and the comfort club indices.
192 T.K. Giap et al.

The GLCI overlaps with the comfort club in six cities: Auckland (2), Copenhagen
(1), Geneva (1), Helsinki (1), Melbourne (1) and Zurich (1). The number in the
parenthesis reports the number of times that this GLCI top-10 city appears in the
comfort club’s top-10 lists. Once again, the amount of overlap between GLCI and
the comfort club indices greatly exceeds the overlap between the clout club and the
comfort club.
Since the GLCI fits in both the clout club and the comfort club, we describe it as
belonging to the ‘balanced club indices’. The GLCI is, in short, a more balanced indicator
than the other six city indices. The GLCI balances the emphasis of the clout club indices
on a city’s ability to project influence and to provide economic opportunities against the
emphasis of the Comfort Club Indices on a city’s capacity to delight the aesthetic senses
and to provide recreational activities. This balancing aspect of GLCI can be seen in that it
ranks the top four cities in the Mori, Foreign Policy and Knight Frank indices
substantially lower. GLCI ranks New York as 17th, London as 22nd, Paris as 13th, and
Tokyo as 18th – and this ranking is, in turn, substantially better than the ranking by the
Mori Index, which puts New York as 47th, London as 38th, Paris as 30th, and Tokyo as
46th.
A Freudian analogy could be usefully employed to explore the nature of the GLCI. In
a serious sense, the GLCI is the ‘ego’ that combines the ‘super-ego’ of the clout club and
the ‘id’ of the comfort club. In Freudian analysis, the super-ego is the critical and
moralising function that operates according to the perfection principle; the id is the set of
unorganised instincts that operates according to the pleasure principle; and the ego is the
organised part of the psyche that operates according to the reality principle. In accordance
with the reality principle, the ego is constantly balancing the demand for absolute
adherence to the social ideals of the super-ego (being driven by the perfection principle)
with the demand for instant gratification of individual desires of the id (being driven by
the pleasure principle). The super-ego concept is analogous to the hardnosed emphasis of
the clout club on money and influence; the id concept is analogous to the feely-touchy
emphasis of the comfort club on pleasant living; and the ego concept is analogous to the
‘middle way’ of the GLCI methodology.
We would like to go further and suggest that we can deduce a particular ethical bias
in the GLCI ranking from the degree of attractiveness that the GLCI ranking has for
different audiences. Table 11 reports the top-20 city rankings of different types of
individuals as measured by the Knight Frank Index and the Mori Index. Part A of
Table 11 divides personalities into four types – the ultra-high net worth individual
(UHNWI), the entrepreneur, the hedonist, and the romantic – and gives the top 20 city
ranking for each personality type. Part B [from Mori Memorial Foundation (2011,
Figures 1–4 and Tables 7–8)] partitions personality types into the manager, the
researcher, and the artist; and reports the top 20 cities for each group.
Of these seven personality types, we would group the researcher, the manager, the
UHNWI, and the entrepreneur types into the super-ego category; and the artist, the
hedonist, and the romantic into the id category. When we match the GLCI ranking with
the city preference ranking of each personality type in Table 11, we find that the GLCI
has 13 matches with the researcher type, 12 matches with the manager type, 11 matches
with the UHNWI type, and four with the entrepreneur type; and ten matches with the
artist type, seven matches with the Hedonist type, six matches with the Romantic type.
 A new approach to measuring the liveability of cities 193

The average number of matches in the super-ego category is 10, and the average
number in the id category is almost 8. This finding reveals that the GLCI is closer to the
clout club indices than to the comfort club indices, but the bias is not large. For example,
the spontaneous artist type still has a 50% match compared with the 65% match of the
highly analytical researcher and the 60% of the hardnosed manager type. The GLCI, in
short, is most suited for identifying preferred cities for residence by ambitious, task-
focused professionals who also value pleasant living; but would not accept the latter in
the absence of good career opportunities. In short, the ethical values embodied in the
GLCI could be described as the maintenance of a balance between work and play, with
work coming before play; and a consistency between responsible global citizenship and
active local actions for environmental sustainability.

Table 11 The world’s top 20 cities for different types of people

Part (A) according to the Knight Frank Index

The ultra-high net
worth individuals The entrepreneur The Hedonist The Romantic
(UHNWIs)
1 New York Shanghai New York Paris
2 London Hong Kong Hong Kong New York
3 Hong Kong Beijing Tokyo London
4 Singapore New York Paris Rome
5 Beijing Mumbai London Tokyo
6 Shanghai Singapore Shanghai Sydney
7 Tokyo London Rio Shanghai
8 Paris Sao Paulo Barcelona Hong Kong
9 Geneva San Francisco Sydney San Francisco
10 Zurich Palo Alto Dubai Vancouver
11 Washington DC Dubai Bangkok Rio
12 Dubai Rio Beijing Venice
13 Mumbai Moscow Singapore Las Vegas
14 Berlin Sydney Rome Buenos Aires
15 Sydney Delhi Las Vegas Barcelona
16 Moscow Istanbul Monaco Istanbul
17 San Francisco Jakarta Vancouver Beijing
18 Los Angeles Lagos San Francisco Dubai
19 Vancouver Dallas Prague Milan
20 Sao Paulo Bangalore Miami Miami
Source: Reproduced from Frank (2011b, p.62)
194 T.K. Giap et al.

Table 11 The world’s top 20 cities for different types of people (continued)

Part (B) according to the Mori Index

The manager The researcher The artist Memo item: GLCI
1 London New York Paris Geneva
2 Singapore Tokyo London Zurich
3 Hong Kong London New York Singapore
4 New York Paris Tokyo Copenhagen
5 Beijing Boston Berlin Helsinki
6 Paris Seoul Vienna Luxembourg
7 Shanghai Singapore Los Angeles Stockholm
8 Tokyo Los Angeles Amsterdam Berlin
9 Zurich San Francisco Madrid Hong Kong
10 Geneva Hong Kong Milan Auckland
11 Amsterdam Sydney San Francisco Melbourne
12 Copenhagen Chicago Beijing Sydney
13 Seoul Berlin Osaka Paris
14 Vancouver Vancouver Chicago Vancouver
15 Vienna Osaka Copenhagen Amsterdam
16 Berlin Amsterdam Brussels Osaka-Kobe
17 Frankfurt Zurich Toronto New York
18 Sydney Geneva Sydney Tokyo
19 Toronto Beijing Vancouver Los Angeles
20 Taipei Vienna Frankfurt Philadelphia
Source: Reproduced from Frank (2011b, p.62)

5 Conclusions

The existing major city indices can be divided into two groups:
1 those that value highly the cities with economic-financial prowess, and strong global
agenda-setting power in political and cultural matters
2 those that value highly the cities with pleasant living in mild climate, scenic
locations.
The GLCI has been able to combine these two aspects by focusing more on the
multi-dimensional needs of the working professionals.
The implicit ethical values of a balance between work and play, and of a balance
between thinking globally and acting locally are values which we are comfortable in
advocating to any city, and which we think most people could accept. By spelling clearly
the construction of GLCI and by selecting a wide range of indicators that are moderately
easy to access, we have enabled a city that wishes to promote this kind of lifestyle to
achieve it by investing in the areas identified as the low-score components of its GLCI
 A new approach to measuring the liveability of cities 195

ranking. Our simulation exercise confirms the feasibility of doing so. In short, our GLCI
can be used to enable the fulfilment of the primal human desire for self-improvement.
Our ranking of the liveability of global cities is necessarily a work-in-progress, even
if it might arguably be the best-in-its-class because of its balanced approach. There have
been some data limitations that we have not yet been able to overcome, and this has
forced us to work at this point with a sample of only 64 global cities. In addition to
working to expand the number of cities covered, we are also working to improve our
methodology by incorporating additional dimensions of liveability into our theoretical
framework, and by searching for better proxies for the variables in the empirical
framework.
Furthermore, the realities of, one, that the global environmental conditions could
change drastically sometimes; and, two, that city administrations and national
governments could move comprehensively to a new socio-economic-political policy
regime occasionally mean that any ranking of cities on their liveability captures only their
relative positions at a particular point in time. For example, the rapid growth of the
Chinese economy has generated substantial resources that have allowed many Chinese
cities to build 21st Century infrastructure in transportation and to undertake ambitious
environmental restoration that would eventually improve the rank of some Chinese cities
significantly. But for the moment, these projects have not yet reached critical mass and
hence have not improved liveability in these cities substantially enough to boost their
ranking. So the rank of a city today is not necessarily a good indicator of its rank in the
future. While the GLCI adopts the maximum entropy principle (actually ‘maximum
agnosticism principle’) by putting equal weights on every category, obtaining the weights
from a survey on people’s preferences would be a more interesting and appropriate
exercise for future research.

References
A.T. Kearney (2010) The Urban Elite: The A.T. Kearney Global Cities Index 2010 [online]
http://www.atkearney.com/images/global/pdf/Urban_Elite-GCI_2010.pdf (accessed 2012).
Economist Intelligence Unit (EIU) (2011a) Hot Spots: Benchmarking Global City Competitiveness
[online] http://www.citigroup.com/citi/citiforcities/pdfs/hotspots.pdf (accessed 2012).
Economist Intelligence Unit (EIU) (2011b) A Summary of the Liveability Ranking and Overview,
August [online] http://www.eiu.com/public/topical_report.aspx?campaignid=liveability2011
(accessed 2012).
Frank, K. (2011a) The Wealth Report: A Global Perspective on Prime Property and Wealth
[online] http://www.knightfrank.com/wealthreport/2011/images/brochure.pdf (accessed 2012).
Frank, K. (2011b) Tales of the Cities: Results of the Knight Frank Global Cities Survey [online]
http://www.knightfrank.com/wealthreport/2011/global-cities-survey/ (accessed 2012).
Glaeser, E. (2011) Triumph of the City: How our Greatest Invention Makes US Richer, Smarter,
Greener, Healthier, and Happier, The Penguin Press.
Ley, A. and Newton, P. (2010) ‘Creating and sustaining liveable cities’, in Kallidaikurichi, S. and
Yuen, B. (Eds.): Developing Living Cities: From Analysis to Action, World Scientific,
Singapore.
Luque-Martinez, T. and Munoz-Leiva, F. (2005) ‘City benchmarking: a methodological proposal
referring specifically to Granada’, Cities, Vol. 22, No. 6, pp.411–423.
Lyndhurst, B. (2004) Liveability & Sustainable Development: Bad Habits & Hard Choices, July,
Final Report for the UK Office of the Deputy Prime Minister.
196 T.K. Giap et al.

Mercer (2011) The 2011 Quality of Living Worldwide City Rankings – Mercer Survey [online]
http://www.mercer.com/qualityoflivingpr#city-rankings (accessed 2012).
Mori Memorial Foundation (2011) Global Power City Index 2011: Summary [online]
http://www.mori-m-foundation.or.jp/english/research/project/6/pdf/GPCI2011_English.pdf
(accessed 2012).
Organization for Economic Cooperation and Development (OECD) (2007) Competitive Cities
Report, OECD, Geneva.
Sassen, S. (1991) The Global City: New York, London, Princeton University Press, Tokyo.
Tan, K.G., Woo, W.T., Tan, K.Y., Low, L. and Aw, E.L.G. (2012) Ranking the Liveability of the
World’s Major Cities: The Global Liveable Cities Index (GLCI), World Scientific, Singapore.
United Nations (2011) World Urbanization Prospects, The 2011 Revision, United Nations, Geneva.
University of British Columbia (UBC) (2012) Taking the Earth’s pulse: UBC Scientists Unveil a
New Economic and Environmental Index, Media Release on February 20, 2012.
Urbis (2008) Enhancing Victoria’s Liveability, Victorian Competition and Efficiency Commission,
Melbourne.
Vuchic, V. (1999) Transportation for Livable Cities, Center for Urban Policy Research,
New Jersey.
Woo, W.T. (2007) ‘The challenges of governance structure, trade disputes and natural environment
to China’s growth’, Comparative Economic Studies, Vol. 40, No. 4, pp.572–602.
Woolcock, G. (2009) ‘Measuring up? Assessing the liveability of Australian cities’, State of
Australian Citis: National Conference, Promaco Conventions.

Notes
1 This is why one measure of a country’s efforts to increase sustainability that we use in the
empirical is (G/GDP) rather than the commonly-used G or (G/P).
2 The reader interested in the environmental challenges of China could find a brief review in
Woo (2007).
3 Each of the five main categories comprises of a different number of indicators, subjecting the
weakest 20% indicators to category limitations or setting an equal number of indicators for
each category would introduce biasness into the weightings of the simulation.
4 The top 10 cities in GLCI actually turn out to be 11 cities because Auckland and Melbourne
compete for the tenth place together.