review paper

lable at ScienceDirect

Habitat International 45 (2015) 3e9

Contents lists avai

Habitat International

journal homepage: www.elsevier.com/locate/habitatint

A framework for ‘City Prosperity Index’: Linking indicators, analysis and policy

Cecilia Wong*

Centre for Urban Policy Studies, School of Environment, Education and Development, The University of Manchester, Manchester M13 9PL, United Kingdom

a r t i c l e i n f o

Article history: Available online 22 July 2014

Keywords: City Prosperity Index UN-Habitat Indicators Urban planning Spatial dynamics Urban change

* Tel.: þ44 161 275 0680. E-mail address: [email protected].

http://dx.doi.org/10.1016/j.habitatint.2014.06.018 0197-3975/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

This paper argues for a more robust and flexible framework to develop the ‘City Prosperity Index’ (CPI), one which is able to connect indicators and analytical intelligence with the policy needs of urban planners and government strategists. The adoption of a more progressive and balanced agenda of ‘people-centred’ urban prosperity in the UN-Habitat's newly developed CPI has already led to a more holistic approach to integrating productivity, infrastructure, quality of life, equity and social inclusion, and environmental sustainability into a coherent framework. Building on this international agenda, there is still scope to critically revise and improve the conceptual and methodological framework of the CPI, probably in an incremental manner, to make it a more tailored policy instrument that can truly address the different sets of challenges faced by cities in different regions under different socio-spatial contexts to achieve sustainable prosperity.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

The measurement and use of quantitative indicators is closely intertwined with the prevailing policy regime. It is because of this pragmatic purpose that the policy concept to be measured is not static but undergoes a dynamic process of problem definition (Greer, 1969). Over the years, supranational organisations have shown interest in assessing the state of urban development across different nations. The World Bank (2013) has compiled the annual world development indicators series since 1960 to monitor the achievement towards international development goals. Since 1990, the United Nations (2013) has published the Human Development Index (HDI) to rank countries into four levels of human develop- ment. Since early 2000s, Eurostat has periodically developed ‘Ur- ban Audits’ to improve the European Commission's knowledge of quality of life in urban areas across Europe (EC, 2000). The Asian Development Bank has also endeavoured to compile the Cities Data Book to inform the management of the urban sector in Asia (Westfall & de Villa, 2001).

Recent international policy discourse has moved away from national and regional perspectives to focus on cities as drivers of the growth agenda. This shift is closely related to the unstoppable pace of urbanisation in developing countries. According to the 2011 United Nations figures, over two-thirds of the world population is

forecast to be urban dwellers by 2050; though the urbanisation level varies in different regions and largely associates with their development levels (UNDESA, 2012). The latest proposal to develop a City Prosperity Index (CPI) by the UN-Habitat (2012: iv) came at ‘a time of crisis’, as introduced by Joan Clos, the Executive Director of UN-Habitat. The global financial crisis and the challenges brought by climate change call for drastic solutions; cities are seen as the driver to move from such crises towards the prosperity pathway.

The critical questions to be asked are: (1) what is distinctive about this new CPI; (2) what are the methodological and concep- tual challenges in developing a robust index to measure city prosperity; and (3) what are the key parameters for developing a robust and flexible indicator framework to measure and bench- mark urban prosperity? This paper aims to address these key questions and, more importantly, to identify the key underpinning principles to improve the CPI framework to connect indicators and analytical intelligence with the policy needs of urban planners and government strategists.

Following this introduction, the paper will first highlight the key characteristics and provide a critical appraisal of the CPI presented in the State of the World's Cities 2012/13 report (SoWC) (UN-Habitat, 2012). It will then identify some key conceptual and methodolog- ical challenges to be addressed for future refinement of the CPI. The fourth section will discuss the principles and ground rules that can potentially be used to underpin the development of a robust and policy-oriented indicator framework for the CPI by drawing from international experience on indicators research. The final section

C. Wong / Habitat International 45 (2015) 3e94

concludes by drawing out some wider methodological and practical implications for the measurement of urban prosperity.

The wheel of prosperity: an integrative and progressive agenda

The conception of the CPI comes with a strong assertion of the vitality and transformative dynamics of cities and thus their importance as ‘the world moves into the urban age’ (UN-Habitat, 2012: v) for a new type of city ‘that is a “good”, people-centred … shedding off the inefficient, unsustainable forms of functionalities of the city of the previous century’ (UN-Habitat, 2012: iv). These inspirational qualities, in many ways, resemble the underpinning rationale of the HDI to ‘shift the focus of development economics from national income accounting to people centred policies’ (ul Haq, 1995 quoted in Fukda-Parr, 2003: 302) by measuring life ex- pectancy, adult literacy rate, GDP and purchasing power parity (UN, 2013). The CPI sets out with a strong critique of the ‘GDP fetishism’ and argues for the need to move towards measuring the broader conception of human and societal well-being.

By putting prosperity within a ‘people-centred’ agenda, UN- Habitat advocates its own approach by defining a prosperous city as one that possesses these essential qualities:

� Productivity: contributes to economic growth and development;

� Infrastructure: deploys infrastructure, physical assets and amenities;

� Quality of life: provides social services for improved living standards and guarantees safety and security;

� Equity and social inclusion: ensures equitable distribution of wealth and benefits and eradicates poverty; and

� Environmental sustainability: protects the urban environment and preserves natural assets while creating wealth.

These five dimensions of prosperity (see Fig. 1) are regarded as the spokes of the wheel of prosperity, each of which is measured by

Fig. 1. The wheel of urban prosperity. Source: UN-Habitat (2012), Figure 1.1 (p.15).

a number of indicators or sub-indices. The operational definitions of the five dimensions of prosperity are provided in Table 1. It is important to note that, rather than developing a total new set of indicators, those in the HDI are used to re-calculate the productivity and quality of life components of the CPI. This will allow continuity and cross-comparison between the HDI at the national level and the two components of the CPI at the city level.

One can argue that such an all-embracing definition of pros- perity makes it apply to everything and therefore difficult to deliver. Some may even argue that the broadened notion of pros- perity by the UN-Habitat may strengthen economic growth as an acceptable international policy discourse amidst the global eco- nomic crisis. Apart from advocating a more balanced and integra- tive framework for measuring prosperity, the UN-Habitat places ‘government, institutions, laws and urban planning’ at the centre to achieve sustainable prosperity. Recognising that it will be difficult for any city to achieve all five aspects on an equal basis at any point in time, policy intervention is thus required to restore the balance. The use of a wheel provides a graphic analogy (see Fig. 1): the hub of the wheel represents urban power and innovative policies, which provides balance and structure to the five inter-dependent prosperity spokes to drive forward along the road of ‘prosperity’. Apart from law and government institutions, urban planning is seen as vital in providing a dynamic hub of city development. Urban planning provides an integrative perspective by overcoming sec- toral policy silos and provides an inclusive agenda by balancing the nexus between different dimensions of urban functionalities (Wong & Watkins, 2009).

Another distinctive development in the SoWC report is the strong articulation of cities as engines of prosperity. Both the World Bank and the United Nations focused on the national level when compiling composite indices to benchmark performance in the past. The agenda of seeking spatial equality has shifted to the urban level, which signifies the importance of adopting more localised and contextualised approaches of growth and development. The notion that global challenges have to be met with local responses has singled cities out as ‘ready, flexible and creative platforms’ to provide remedy in ‘a pragmatic, balanced and efficient way’ (UN- Habitat, 2012: 11).

This discussion shows that there are some very positive features of the CPI. However, it is important to question the extent to which it addresses the pitfalls of using aggregate indices to measure development. There are other urban development indices pro- duced by the European Commission, the Asian Development Bank and the United Nations. Despite these efforts, our understanding of urban change is still limited in three ways: firstly, most reports tend to have partial global geographic coverage of specific regions (e.g.

Table 1 City Prosperity Index: definition and variables.

Productivity index It measures the total output of goods and services (value added) produced by a city's population during a specific year by including variables such as capital investment, formal/informal employment, inflation, trade, savings, export/import and household income/ consumption.

Quality of life index It is a combination of three sub-indices: education, health and public space.

Infrastructure development index

It combines two sub-indices: infrastructure proper and housing.

Equity and social inclusion index

It has three sub-indexes: air quality (PM10), CO2 emissions and indoor pollution.

Environmental sustainability index

It combines measures of inequality of income/ consumption (Gini coefficient) and inequality of access to services and infrastructure.

Source: UN-Habitat, 2012: 14.

C. Wong / Habitat International 45 (2015) 3e9 5

Asian Cities Data Book); secondly, many tend to focus on ranking by measuring performance at the national level (e.g. HDI); and thirdly, most provide a snapshot of a particular aspect of social change. The launch of the CPI does seem to address these issues by creating an all-embracing index to measure the urban prosperity of cities across the globe. This, in many ways, represents a step-change of how we benchmark and compare urban prosperity. While the principles and measurement framework are identified in the SoWC report, there are still significant knowledge gaps in the framing and operationalization of ‘prosperity’. The next section tries to dissect the conceptual and methodological challenges of making CPI more than just a measurement tool to inform policy-making under different socio-spatial contexts.

Conceptual and methodological challenges: what is in store?

Urban change is a continuous process of spatial transformation. Global financial flows and the use of telecommunications tech- nologies have altered the international economic development landscape in which cities compete. At the same time, long-standing economic decline and social problems in urban areas have triggered labour market restructuring and other socio-cultural adjustments. The outcomes of the process of change have been mixed and are contingent upon the endowment and exploitation of assets and resources of the urban areas and their wider functional hinterlands.

(1) Openness of the concepts and their complex interactions

The use of a multi-dimensional wheel to integrate productivity, infrastructure, quality of life, equity and social inclusion, and environmental sustainability within the overarching concept of prosperity shows the commitment to achieve a more holistic form of urban prosperity and development. The five dimensions are themselves fuzzy and open concepts, subject to very different un- derstanding and interpretations (Wong, 2006) and will require trial and error to make them workable (Innes, 1990). It is the complexity of the development processes of the five dimensions, and the fuzziness of these concepts, that make it a difficult task to establish their inter-relationships. However, it is more important to under- stand how the outcome of sustainable prosperity can be achieved, and why it is happening, than simply to measure it.

Concepts such as environmental sustainability and social exclusion encapsulate both the process of change and the state of development. The operation of different aspects of prosperity may reinforce and enhance the restructuring process of cities (such as quality of life and productivity), but their interaction can also be contentious (such as achieving economic growth and a sustainable environment) (Wong & Watkins, 2009). For example, the causal relationship between quality of life and economic prosperity is a difficult and controversial topic. Castells (1989: 52) regards quality of life as the result of the characteristics of the high tech industry in Silicon Valley rather than the determinant of its location pattern. Findlay, Rogerson, and Morris (1989) also fail to find any significant correlation between the Quality of Life Index and the Local Pros- perity Index in British cities. Wong (2001) finds that quality of life is important to local economic development provided that the traditional factors of production are already in place. Quality of life was found to be more important in shaping the reproduction space than the production space of cities. Likewise, Boddy (1999) finds that a lack of social cohesion may not impede competitiveness; cities that ‘perform’ well still tend to be an obstinate persistence of diverse forms of social exclusion. Rapid economic growth and exclusion can often coexist within the same urban space (Wong et al., 2011). New York is the case in point. Ranked the world's second most prosperous cities (UN-Habitat, 2012: 19), its

performance on the Equity Index was far below the other di- mensions of prosperity.

In addressing complex issues with multiple objectives and interlocking activities like sustainable prosperity, urban planning and policy intervention have to deal with boundaries in a flexible way by addressing ‘polyrationality’ and understanding how cities are socially produced by a wide variety of actors and practices (Davy, 2008). Many of the concepts entangled in the CPI fall into what Rittel and Webber (1973) called ‘wicked problems’ that are embedded in a dynamic social context, which makes each problem unique but also difficult or impossible to solve (Rae & Wong, 2012). The concept of quality of life is a particularly problematic one as the outcome of a people-centred notion of prosperity, to many people, will be synonymous with quality of life.

(2) Quality of data and estimates used to develop the composite indices

Since the variables to compute the productivity and quality of life sub-indices are based on indicators used in the HDI, it is useful to revisit the methodology used to compile those indicators. Ac- cording to Human Development Report (UNDP, 2001: 137), ‘when data are missing for one component, a country will still be included if a reasonable estimate can be found from another source. As a result of revisions in data and methodology over time, the HDI values and ranks are not comparable across editions of the Report’. In general, a wider range of indicators tends to be found at the country level than for units at lower spatial scales. The challenge of getting accurate data at the city level across the world will be much higher. When esti- mated data of different indicators are combined to create sub- indices, the errors and bias involved can be exaggerated. The ef- fects may vary across different sub-indices and geography, as the level of problems encountered in compiling reliable data will vary.

There is a need to understand these methodological issues as they will cumulatively compound the problems that affect the robustness and reliability of different sub-indices and the aggre- gated CPI rankings. This will also limit the ability to conduct change analysis to ascertain the dynamic performance of cities towards achieving sustainable prosperity over time. In order to improve the robustness of the measurement, there are three key questions to be addressed: (a) How to get accurate data at the city level across the world? (b) Will the robustness and reliability of different sub- indices vary? (c) Is there any systematic measurement error against the value computed for cities in any particular regions? Meanwhile, the UN-Habitat should provide more discussion on the methodological limitations and give “health warnings” on the in- terpretations of the rankings and the analysis in the report.

The difficulties encountered in collecting the array of data for cities at different stages of development with very different socio- economic and political circumstances cannot be overlooked. Even in countries with a strong monitoring culture and established data collection practice, such as the USA and the UK, the development of a coherent and reliable set of indicators for urban areas is not a simple task (Wong, 2006). There was a more honest and open discussion in the Cities Data Book project and the problems iden- tified in the Asian exercise are not so different from those faced in the UK (Westfall & de Villa, 2001). The challenges to compile data for a global CPI will be even greater, as a high degree of compati- bility and consistency is more difficult to achieve with a large number of cities in developing countries that have less-developed statistical systems. The cost of collecting the relevant data for the CPI currently falls on the respective city governments and this has proved to be a major problem as many cities have not provided data and are excluded from the analysis. This issue was found to be problematic even in Europe when compiling indicators for Urban

C. Wong / Habitat International 45 (2015) 3e96

Audit II (RWI, DIFU, NEA, & PRAC, 2010) e the response rates from Germany, the Netherlands, Belgium and Luxembourg were high, but very low from the rest of Europe. This links to the question of how often such a data collection exercise should be carried out to update the CPI. There is an obvious trade-off between the frequency and cost of data collection. These more pragmatic issues are not addressed by the UN-Habitat other than the fact that the CPI will be developed in an incremental manner.

(3) Weighting and analytical methods

The development of composite sub-indices and an overall index tend to be the default options used to simplify an indicator set, as it provides a hard and fast technical synthesis (Wong, 2006). The challenge is then how to choose the most robust ‘weighting’ system to combine the indicators. The study by Bagolin and Comim (2008) demonstrates the drastic change in the ranking of the HDI by applying different weighting systems to the indicators. It is thus important to carry out sensitivity analysis (e.g. Coombes, Wong, & Raybould, 1993; Coombes, Raybould, Wong, & Openshaw, 1995) of the assembled database to identify differences in the outcome produced by alternative weighting approaches before making the final judgement. The UN-Habitat has not clearly explained the methodology and the weighting schemes used to create the five prosperity indices and the overall index in the report. It only de- clares that, ‘Although more refinement is still needed in terms of what indicators are included in the index and with which respec- tive weightings, those that have been selected offer the possibility of disaggregating the different dimensions of prosperity, in the process identifying policy intervention areas’ (UN-Habitat, 2012: 18). Since the approach used to derive weighting schemes is always a contentious issue and subject to debate, it would be helpful for the United Nations to open up the discussion on this methodo- logical black box.

(4) Spatial dynamics and definition of cities

The most powerful change in our urban system has been continuing spatial decentralisation. In the developing world, a defining feature of cities ‘is an outward expansion far beyond formal administrative boundaries, largely propelled by the use of the auto- mobile and land speculation. ….. that urban land cover grew, on average, more than double the growth of the urban population’ (UN- Habitat, 2012: 28). In order to harness the potential pool of a pro- fessional workforce and other development resources, the devel- opment of cities has to be seen within the broader spatial context in which they closely interact and connect (Robson, Parkinson, Boddy, & Maclennan, 2000).

This raises the thorny issue of the appropriateness of measuring city prosperity by following administrative boundaries. The prob- lem of using administrative areas is that it may distort the spatial dynamics operating between a city and its wider spatial context. In the under-bounded city (see Fig. 2), the administratively defined city is smaller than the physical urban aggregate, while the oppo- site is true for an over-bounded city (Carter, 1981). The use of administrative boundaries, rather than economic functional areas (for example, labour and housing markets), can give an under- stated/exaggerated impression of urban performance. As illustrated in Fig. 2, with the same spatial patterns of prosperity, the different configurations of district boundaries within the city can lead to very different measurement outcomes. The growth of urban areas beyond their administrative boundaries into surrounding hinter- lands has also led to Castells (2007: 1) proclaiming that ‘the cate- gory (“the city”) has become theoretically and practically obsolete’. This has characterised the development patterns of many Chinese

cities (Cheng, Bertolini, Clercq, & Kapoen, 2013; Tao, Su, Liu, & Cao, 2010; Zhang, 2000). There is thus a need to define cities in a more meaningful way to take into account the urban core and hinterland relationship (Deas & Giordano, 2001) and to avoid the misleading risks caused by the wrong delimitation of areas.

(5) Relationship between urban change and external global forces

The pervasive forces of globalisation act as macro agents to accelerate the restructuring process of complex urban system and have led to spatial inequality of development and stimulated local actors to formulate their own strategies. While the important role of cities as the engine of growth and the centrepiece to address global crises is emphasised in the report, the CPI does not help to disentangle the different driving forces of change or to examine their interactive effects. Urban change can be due to structural changes and historic inertia at the local level, as well as external factors from national and global forces (Rogerson, 1999), or indeed the interaction of internal and external factors. The SoWC report provides some analysis on the associated performance between national and city levels, but it is not sufficient to disentangle the interactive effects across multiple spatial scales.

The application of statistical techniques, such as principal component analysis in Urban Audit II (RWI et al., 2010), to explore the relationships of the full indicator set and to detect the under- lying dimensions of different aspects of prosperity (Wong, 2002) would be a worthwhile exercise. Likewise, in the Cities Data Book (Westfall & de Villa, 2001), different analytical methods were used to carry out crosscutting, thematic analysis and benchmarking of urban performance across the 18 Asian cities. It is important to develop rigorous analyses to underpin the nature of interactions between the different prosperity dimensions of the wheel and to validate the robustness of the five-fold classification. It will remain important to monitor issues at different spatial levels within a coherent framework if the CPI is to be truly relevant for city man- agement. This also links to the previous argument over the problem of using convenient city administrative boundaries. Administrative boundaries do not necessarily reflect functional areas in terms of social, economic and environmental linkages. There is, therefore, a need to build up monitoring capability to organise, analyse and display data at varying spatial scales.

(6) Tracking progress and policy-making

In order to inform urban planning and to develop an institu- tional framework, it is important to interpret the degree of progress made by a city. The use of benchmarking provides a yardstick to gauge the relative performance of a city by assessing its progress and achievement against other comparator areas or the national average. However, the use of the HDI to produce national rankings has been much criticised for its failure to provide more nuanced and contextualised interpretation of such rankings (Bagolin & Comim, 2008). It is impossible to establish general non- contextual laws since different areas perform under very diverse sets of socio-economic and political circumstances. This strongly suggests that the comparators should be among areas with similar circumstances. In order to confront this dilemma, a structur- eeperformance model was proposed (Carlisle, 1972) to analyse the differential socio-economic contexts (structure) against which the urban area performs (performance). Such an analytical framework does not aim to develop a causal model, but will provide a distinction between the more descriptive nature of the socio- economic and historical conditions (structure) and the goal and outcome-oriented performance measures (performance). However, due to the complexity and the intertwining of different socio-

Fig. 2. Spatial boundaries and indicator values. Source: author.

C. Wong / Habitat International 45 (2015) 3e9 7

economic issues, it is impossible to operationally untangle the web of conditions and outcomes (Wong, 2002; Wong et al., 2004).

The use of benchmarking aims to assess relative performance, should it be economic growth under a buoyant economy or the resilience to economic decline during a period of recession, by taking into account their operating circumstances as well as the external forces of change. Taylor (2000) argues that, whilst benchmarking against each other and against future improvements is valid, this process can be exclusive and distort attention if simply focussing on particular negative aspects of urban problems. The publication of rankings will lead to the production of league tables and, in some cases, led to the stereotyping of particular cities and regions. This suggests that when analysing indicators, one has to be sensitive to the presentation and dissemination of the findings (Wong, 2006). Composite indices have the advantage of avoiding information overload, but they tend to conceal the detailed infor- mation on the diversity of performance over different aspects of urban change. While the UN-Habitat argues that the CPI allows urban authorities to have a handle on balancing different aspects of prosperity, the use of composite sub-indices rather than displaying individual indicator values has significantly reduced the amount of policy intelligence.

Analytical principles of a policy-oriented indicator framework

The conceptual and methodological challenges of developing the CPI into a robust policy instrument are not easily resolved within a short period of time. The UN-Habitat has rightly adopted a gradual approach to developing the CPI. However, the SoWC report mainly mentions refinement with respect to indicators and weighting systems. The discussion above suggests that the chal- lenges are more than a technical fix on data and weighting schemes, but it should involve further theoretical thinking over the conceptualisation of different aspects of prosperity; the causal relationship between different driving forces operating at multi- spatial levels; the pragmatic concerns associated with data collec- tion responsibility and costs; and capacity building with

stakeholders to ensure the development of CPI as a policy relevant instrument.

One major ingredient missing from the SoWC report is a clear conceptual understanding of the overarching indicator framework and the analytical principles that underpin the development of this framework. This is particularly important if the CPI is to be devel- oped and adjusted over time for use as an urban management tool. Based on previous research on developing indicators to inform urban planning and local development (Rae & Wong, 2012; Wong, 2006; Wong & Watkins, 2009), a set of key analytical principles are proposed for the future development of CPI. The seven proposed principles are:

1. Consistency and comparability: data have to be collected on a common spatial and temporal basis, under a clearly identified set of definitions for both the indicators and the spatial defini- tion of cities to allow meaningful analysis. The SoWC report is unclear about the eligibility criteria for including or excluding a city to be involved in the study. There is obvious political sensitivity in dealing with such issue. However, ducking it rather than working with stakeholders to find a satisfactory solution will not help to achieve the promise made in the report.

2. Tracking progress and change: the analysis of the indicators and sub-indices of different aspects of the CPI should provide clear narration of the nature and direction of progress and should not be subject to open interpretation. There is a need to ensure that the indices are constructed in a way that will allow policy- makers and stakeholders to track changes made in different aspects of prosperity to steer policy interventions.

3. Benchmarking and cross-comparison: meaningful interpreta- tion of a city's indicator values can be enhanced by making reference to the performance of other cities that share similar socio-economic and political conditions. The idea of a struc- tureeperformance model should be considered to establish the baseline contextual conditions of all cities. Those with similar circumstances could then be grouped together for comparison over their prosperity performance. This means that the analysis

C. Wong / Habitat International 45 (2015) 3e98

should compare and contrast cities with similar conditions rather than their overall CPI ranks.

4. Multi-units of analysis: due to the complexity of urban change, it is important to develop a multi-spatial framework to provide a flexible analytical structure for assessing city performance against the wider national and regional contexts. The report has already provided some analysis at the national and regional levels, but they are not consistent and systematic. A fully developed multi-spatial level analysis would help to develop more informed policy intelligence.

5. Exploration of co-variations and interactive effects: there is a need to provide sufficient details across different aspects of urban prosperity as well as including contextual information to facilitate the analysis of co-variations and interactive effects across different dimensions of city prosperity. The idea of using principal component analysis to perform some exploratory analysis could be a very good starting point.

6. Use of soft indicators and qualitative information: while the CPI is largely based on quantitative data, the analysis of the indicator and index values will require the use of qualitative data and other soft information to assess progress or to enrich in- terpretations. While brief comments were made in the SoWC report for certain cities, it would be very helpful to have sys- tematic commentaries on each city. It would be useful to develop a pen picture for each city and their associated spatial contexts at national and global regional levels, complemented by succinct qualitative commentaries.

7. A communicative and learning framework: to embed the CPI into urban planning and policy making processes will require the participation of a wide range of urban partners and stakeholders. This means that the development of the CPI in- dicator framework has to focus on the communication of policy intelligence that can be shared among key partners and stakeholders to facilitate policy learning and debate. Web- based interactive visualisation tools will be a way forward (Kingston, 2007; Wong, Baker, Hincks, Schulze B€aing, & Webb, 2012), as different level of analysis, pen pictures, commen- taries, charts and graphs and indicators can be provided in one single website to allow users to manipulate and develop their own analysis.

The first three of these principles are inter-related in that the achievement of one will facilitate the development of the others. These are, therefore, seen as more critical to firmly establish the quality threshold of the CPI as a robust and relevant measure of urban prosperity. The fourth, fifth and sixth principles are related to more in-depth methodological development of the analysis. Since the CPI is a huge data collection exercise, it is important for the UN- Habitat to extract the best policy intelligence out from the data to inform policy thinking. However, there will be major methodo- logical challenges and costs involved. The last principle is about partnership and capacity building, which is politically challenging and will take time for different stakeholders to engage in the debate. More importantly, success in gaining political support and the momentum necessary to develop a bona fide CPI will help to overcome some of the data collection and methodological challenges.

Conclusion

The emphasis on urban prosperity provides an interesting platform for urbanists to rethink our place-making approaches and the need to put people back to the centre of the conception. However, this discussion highlights the methodological, conceptual and political challenges ahead.

Since the methodological details provided in the SoWC report are rather limited, it is difficult to make a fully robust assessment of the CPI. It is thus important for the UN-Habitat to publish the precise definition of different elements of CPI to elicit comments and reviews from experts and stakeholders. In addition, there is a need to clarify how a city is defined in the measurement of CPI and to inject more rigour in the conceptualisation of the driving forces, different dimensions and the outcomes of prosperity.

The analytical principles set out above aim to provide a basis to inform the debate over what might be the most effective moni- toring framework of international urban prosperity over time. It is important to note that no single set of indicators will ever be optimal, and the identified indicators by the UN-Habitat will inevitably change when new policy issues emerge and better quality data sets come along. It is therefore important to develop a set of guiding principles to provide some parameters and flexibility to establish future indicator frameworks that have utility in a va- riety of different contexts. It is also important for stakeholders in each region to take this framework forward and to share re- sponsibility for its modification and refinement.

References

Bagolin, I. P., & Comim, F. V. (2008). Human Development Index (HDI) and its family of indexes: an evolving critical review. Revista de Economia, 34(2), 7e28.

Boddy, M. (1999). Geographical economics and urban competitiveness: a critique. Urban Studies, 36(5/6), 811e842.

Carter, H. (1981). The study of urban geography (3rd ed.). London: Edward Arnold. Carlisle, E. (1972). The conceptual structure of social indicators. Published for the

Social Science Research Council. In A. Shonfield, & S. Shaw (Eds.), Social in- dicators and social policy. London: Heinemann Educational Books.

Castells, M. (1989). The information city. Oxford: Blackwell. Castells, M. (2007). Why the megacities focus? Megacities in the new world dis-

order. In J. Perlman, & S. O'Meara (Eds.), The state of the world e Our urban future (pp. 1e16). E-publishing http://www.megacitiesproject.org.

Cheng, J., Bertolini, L., Clercq, F. L., & Kapoen, L. (2013). Understanding urban net- works: comparing a node-, a density- and an accessibility-based view. Cities, 31, 165e176.

Coombes, M., Raybould, S., Wong, C., & Openshaw, S. (1995). The 1991 deprivation index: a review of approaches. London: HMSO.

Coombes, M., Wong, C., & Raybould, S. (1993). Local environment index: Infra- structural resources dimension, a final report to the Department of Employment, Centre for Urban and Regional Development Studies. Newcastle upon Tyne: University of Newcastle upon Tyne.

Davy, B. (2008). Plan it without a condom! Planning Theory, 7, 301e317. Deas, I., & Giordano, B. (2001). Conceptualising and measuring urban competi-

tiveness in major English cities: an exploratory approach. Environment and Planning A, 33(8), 1411e1429.

European Commission. (2000). The urban audit: Towards the benchmarking of quality of life in 58 European Cities. In The Yearbook (Volume 1). Luxembourg: Office for the Official Publications of the European Communities.

Findlay, A., Rogerson, R., & Morris, A. (1989). In what sense ‘indicators’ of quality of life? Built Environment, 14(2), 96e106.

Fukuda-Parr, S. (2003). The human development programme: operationalizing Sen's ideas on capabilities. Feminist Economics, 9(2e3), 301e317.

Greer, S. (1969). The logic of social inquiry. Chicago: Aldine. ul Haq, M. (1995). Reflections on human development. New York: Oxford University

Press. Innes, J. (1990). Knowledge and public policy: The search of meaningful indicators.

New Brunswick: Transaction Publishers. Kingston, R. (2007). Public participation in local policy decision-making: the role of

web-based mapping. The Cartographic Journal, 44(2), 138e144. Rae, A., & Wong, C. (2012). Monitoring spatial planning policies: towards an

analytical, adaptive and spatial approach to a ‘wicked problem’. Environment and Planning B: Planning and Design, 39, 880e896.

Rittel, H., & Webber, M. (1973). Dilemmas in a general theory of planning. Policy Sciences, 4, 155e169.

Robson, B., Parkinson, M., Boddy, M., & Maclennan, D. (2000). The state of English cities. London: Department for Environment, Transport and the Regions.

Rogerson, R. J. (1999). Quality of life and city competitiveness. Urban Studies, 36(5e6), 969e985.

RWI (Rheinisch-Westf€alisches Institut für Wirtschaftsforschung), German Institute of Urban Affairs e DIFU, NEA Transport Research and Training, & Policy Research and Consultancy e PRAC. (2010). Urban audit II: Second state of Eu- ropean cities report. Research Project for the European Commission, DG Regional Policy. Essen, Germany: RWI.

C. Wong / Habitat International 45 (2015) 3e9 9

Tao, R., Su, F., Liu, M., & Cao, G. (2010). Land leasing and local public finance in China's regional development: evidence from prefecture-level cities. Urban Studies, 47(10), 2217e2236.

Taylor, M. (2000). Communities in the lead: power, organisational capacity and social capital. Urban Studies, 37(5), 1019e1035.

UNDESA. (2012). World urbanization prospects e The 2011 revision. New York: United Nations: United Nations Department of Economics and Social Affairs.

United Nations Development Programme [UNDP]. (2001). Human development report 2001: Making new technologies work for human development. New York and Oxford: UNDP.

United Nations Development Programme [UNDP]. (2013). Human development report 2013-the rise of the South: Human progress in a diverse world. New York: UNDP.

United Nations Human Settlements Programme [UN-Habitat]. (2012). State of the world’s cities 2012/13: Prosperity of cities. Nairobi, Kenya: UN-HABITAT.

Westfall, M. S., & de Villa, V. A. (2001). Urban indicators for managing cities: Cities data book. Manila: Asian Development Bank.

Wong, C. (2001). The relationship between quality of life and local economic development: an empirical study of local authority areas in England. Cities, 18, 25e32.

Wong, C. (2002). Developing indicators to inform local economic development in England. Urban Studies, 39(10), 1833e1863.

Wong, C. (2006). Quantitative indicators for urban and regional planning: The inter- play of policy and methods. In Royal town planning institute library book series. London: Routledge.

Wong, C., & Watkins, C. (2009). Conceptualising spatial planning outcomes: towards an integrative measurement framework. Town Planning Review, 80(4/5), 481e516.

Wong, C., Gibb, K., McGreal, S., Webb, B., Leishman, C., Blair, N., et al. (2011). Housing and neighbourhood monitor 2011: Fragility and recovery. Joseph Rowntree Foundation. Bristol: Policy Press www.hnm.org.uk. For access to the maps and charts in the analysis.

Wong, C., Jeffery, P., Green, A., Owen, D., Coombes, M., & Raybould, S. (2004). Town and city indicators database project. London: the Office of the Deputy Prime Minister.

Wong, C., Baker, M., Hincks, S., Schulze B€aing, A., & Webb, B. (2012). A map for England: Spatial expression of government policies and programmes. London: The Royal Town Planning Institute. http://www.rtpi.org.uk/knowledge/core-issues/ map-for-england/.

World Bank. (2013). World development indicators: 2013. Washington, DC: World Bank.

Zhang, T. (2000). Land market forces and government's role in sprawl: the case of China. Cities, 17, 123e135.