Application of the “Urban Governance Index” to water service provisions: Between rhetoric and reality

Habitat International 49 (2015) 435e444

Contents lists available at ScienceDirect

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

Application of the “Urban Governance Index” to water service provisions: Between rhetoric and reality Luisa Moretto Universit e libre de Bruxelles, Rue Belvedere, 21, 1050 Ixelles, Belgium

a r t i c l e i n f o

a b s t r a c t

Article history: Received 3 January 2015 Received in revised form 22 May 2015 Accepted 12 June 2015 Available online 24 June 2015

When appraising urban governance practices for better service provisions in the Global South, a gap generally emerges between the rhetoric and the reality. Practice-based local governance processes to produce improved urban services are mainly informal and often diverge from the official governance prescriptions and mechanisms for service delivery within the institutional sphere. Here we present the results of an exploration into the complex and uncertain domain of urban governance assessments, focussing on sustainable water supplies for the benefit of the urban poor. By adapting the UN-Habitat Urban Governance Index, we look into the dual dimension of governance to appreciate its formal architecture at the municipal level as well as the nature of informal and locally-specific governance arrangements aimed at improving access to basic services. Water service co-production strategies between public institutions and organised groups of citizens in Venezuela provide an excellent case by which to appraise the two distinct facets of urban governance. We illustrate the limitations of official governance assessment tools in appreciating the extent and vibrancy of local practices and agreements for improved services, and the discrepancies between the normative prescriptions and the governance arrangements on the ground. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Urban Governance Index Governance assessments UN-Habitat Water service provisions Multilevel mixed model design Venezuela

1. Introduction Urban governance has always been a multifaceted and illdefined concept (Burgess, Carmona, & Kolstee, 1997; Halfani, McCarney, & Rodriguez, 1995; Jessop, 2002; Maloutas & Malouta, 2004; Miegville, Ramirez, Ricard, Ropert, & Wagner, 2003; Obeng-Odoom, 2012; Paproski, 1993). But when employed in the Global South, the fuzziness of the urban governance concept is amplified by its two-fold nature. Urban governance obviously presents a formal, institutional, theoretical, and normative aspect, broadly analysed and debated by the international community, and n & unfailingly included in donors' development agendas (Hyde autier, & Mastruzzi, 2004; UNDP, Samuel, 2011; Kaufmann, Le 2007; UN-HABITAT, 2004a; Wilde, Shipra, Laberge, & Moretto, 2009). However, it also displays a more informal, local, communitybased characteristic, which describes how governance works in practice, beyond topedown and external development policies and strategies (Amis et al., 2001; Batley &Moran, 2004; Booth, 2011; Moretto, 2007; Swingedouw, 2005).

E-mail address: [email protected]. http://dx.doi.org/10.1016/j.habitatint.2015.06.004 0197-3975/© 2015 Elsevier Ltd. All rights reserved.

If we look, in particular, to urban governance for improved service provisions, this dual nature is evident. From one side, the concept concerns the sphere of the (new) public management of resources, including the state architecture needed to organise urban service provision and the contractual and/or semi-contractual relationships amongst the stakeholders entering into the urban delivery process (Bakker, 2010; Batley & Larbi, 2004; Burgess et al., 1997; Nickson & Franceys, 2003; Stren, 2012). On the other hand, it also relates to the sphere of the unofficial and unorthodox arrangements that are increasingly being developed at the local level in response to state and market failures in service delivery (Allen, Davila, & Hofman, 2006; Batley & Mcloughlin, 2010; Booth, 2011; Gaventa & Barrett, 2010; Harpham & Boateng, 1997; Joshi & Moore, 2004; Olivier de Sardan, 2009; Wild, Chambers, King, & Harris, 2012). When urban governance for better provisions is appraised, therefore, a gap can occur between different approaches to assessment. As a result, both of these perspectives (formal governance prescriptions and mechanisms or practice-based, informal, local governance agreements and processes) need to be recognised, distinguished, understood and correctly evaluated. They also need to be compared, and mutually verified and integrated, as they both

436

L. Moretto / Habitat International 49 (2015) 435e444

have the same goal: to improve urban living conditions. Here we question whether an urban governance assessment tool able to jointly grasp these two dimensions exists. This is done by systematically overviewing existing governance assessments with respect to their capacity to 1) address cities and, 2) consider the informal and unorthodox side of governance arrangements, with respect to water services in particular, as one of the most important basic services. The UN-Habitat Urban Governance Index is then adapted to the specific case of water service provisions and applied to service co-production in Venezuela. We finally discuss the applicability of this Index, policy implications, and inputs for future research. 2. Urban governance assessments Looking at the huge number of governance indicators and indexes that exists we can easily see that the task of measuring and assessing governance encounters a critical lack of consensus. A quick look into the United Nations Development Programme (UNDP) Global Programme on Democratic Governance Assessments shows that there is no agreement on a single tool or instrument to assess governance. For instance, the UNDP (2007, Wilde et al., 2009) presents in excess of 50 different tools for assessing governance, and these tools are extremely varied. First of all, not only are the actors using governance indicators extremely diverse, including governments, development agencies, NGOs, the media, academic institutions and the private sector (UNDP, 2007), but the users are also varied, as assessment tools are employed to inform citizens on public programmes, to conduct academic research, or to monitor and evaluate governance programmes and projects. Secondly, parts of these assessments are focused on the country level, and are mainly based on statistics and objective measures that include, from time to time, international comparisons of state performance. We can find, for instance, the Country Policy and Institutional Assessment and Governance Matters V, both developed by the World Bank, or the World Governance Assessment by the Overseas Development Institute. But we can also find tools focused on some specific aspects of governance, such as the Bribe Payers Index or the Corruption Perception Index, both produced by Transparency International, or the Press Freedom Index proposed by Reporters without Borders. On the contrary, another subset of these tools is applied to the local level of governance relationships: they usually assess one of the four broad focus areas of governance in a local context (Wilde et al., 2009): representation by local governance (such as the Local Governance Barometer by Impact Alliance), decentralisation processes (the Desde lo local in Mexico, for instance), local democracy (ex., the International IDEA Local Democracy Assessment Guide), and local government (such as the “MIDAMOS” assessment; Wilde et al. (2009) provide many additional examples). These tools usually mix quantitative and qualitative data, often including non-representative sampling and subjective measures such as perception-based evidence collected through surveys and focus groups. Thirdly, concerning geographical coverage, there exist governance assessments that can be applied either worldwide or in some specific regions and countries. Examples of the former are the Human Rights Indicators, developed by the Danish Centre for Human Rights and the Local Integrity Initiative by Global Integrity. In the latter category, which is much more ample, there are tools like the Good Governance for Local Development e GOFORGOLD Index to be used in Afghanistan, the Afrobarometer for Africa, the UNDP Methodological Guidelines for Local Governance, developed for Latin America, and the UNDP Urban Governance Initiative, to be applied in Asia. However, very few of the assessments addressing a more local setting explicitly take into account the urban dimension of governance: in essence, they do not make any particular distinctions

amongst the urban, rural or peri-urban contexts, while proposing a measurement tool for a generic “local government”. Despite the massive production of governance assessment tools, instruments, indicators, and indexes, a measure of urban governance, broadly applicable and also open to local governance arrangements, seems uncommon. Stewart (2006) e while suggesting that the lack of effort directed towards addressing good governance at the urban level rested perhaps on the difficulties related to the data collection process e pointed out two ambitious attempts made at the international level to cope with this shortcoming: the World Bank database concerning globalisation, city governance and city performance (Kaufmann et al., 2004) and the UN-Habitat Urban Governance Index (UN-HABITAT, 2004a).1 The World Bank database aims at exploring whether globalisation matters for sound urban governance and, in turn, whether globalisation and sound urban governance positively affect city performance. This database has been developed covering 412 cities worldwide and relying on 35 variables and indicators belonging to other existing databases.2 To measure city governance, determinants were selected that reveal a positive link between urban governance and the performance of global cities (Kaufmann et al., 2004). This reflects the precise objective of the Bank's database, which is to “investigate empirically what determines the staying power of cities of their performance on a global scale, and whether governance has anything to do it” (Kaufmann et al., 2004: 4). On the other hand, the Urban Governance Index (UGI) has been developed “in order to enable cities to objectively measure the quality of local governance” (Narang, 2005a, 2005b: 1) through a core set of 18 indicators or 25 short-listed indicators, organised in 4 core principles: effectiveness, equity, participation and accountability (UN-HABITAT, 2004c). The objectives of the UGI are two-fold: at a global level it aims “to demonstrate the importance of good urban governance in achieving broad development objectives [while at the local level] the index is expected to catalyse local action to improve the quality of urban governance by developing indicators that respond directly to their unique contexts and needs” (UN-HABITAT, 2004b: 11). Contrary to the World Bank database, selection of the UGI indicators is carried out according to a number of criteria that are also supposed to include informal and unorthodox local governance arrangements, regardless of their relationships with other more market-related global processes, reflecting the differing views on urban governance between the World Bank and the United Nations (Obeng-Odoom, 2012). The UN-Habitat definition of urban governance, in fact, stresses how this concept “includes formal institutions as well as informal arrangements and the social capital of citizens [while claiming that] through good urban governance, citizens are provided with the platform which will allow them to use their talents to the full to improve their social and economic conditions” (UN-HABITAT, 2002: 14). This emphasis on the informality of arrangements and relationships seems to fit comfortably into the variety of urban governance systems based on informal agreements e between formal institutions and practice-rooted community activities e to organise access to basic services. However, if the idea of urban governance upon which the UGI is built appears to be addressing the two facets of this notion, is this Index able to provide a measurement of these two dimensions, and in an integrated way?

1 It is interesting to note that these efforts were undertaken by two leading international aid organisations, coherent with the dominant role that the donor n & Samuel, community has occupied in the governance field since the 1990s (Hyde 2011). 2 The data sources of the World Bank database rely on the UN observatory, EOS database, Taylor database, and KLM database. For further details, see Kaufmann et al., 2004.

L. Moretto / Habitat International 49 (2015) 435e444

437

Fig. 1. Multilevel mixed model design. Source: adapted from Tashakkori and Teddlie (2003). 1 QUAN: indicates a quantitatively-driven project (Morse, 2003) 2 QUAL: indicates a qualitatively-driven project (Morse, 2003)

3. Applying the Urban Governance Index through a multi level mixed model design By considering the UGI indicators, it rapidly appears that, although the Index aims to measure “the quality of relationships and processes between key stakeholders at the local level” (UN-HABITAT, 2004a: 18), in reality it unfortunately seems to fail in assessing the more informal, unrecorded, and practice-based dimension of governance arrangements on the ground. The UGI indicators rely, in effect, on quantitative and factual data and information to be collected at the city level with the support of municipalities and local governments (UN-HABITAT, 2004a; Narang, 2005a, 2005b). While data should be collected through a stakeholder workshop including informal sector representatives (UN-HABITAT, 2004c), this assessment approach can barely encompass the variety and heterogeneity of informal arrangements and practices in water service provisions, often in a multitude of different settlements, that are quite often left as unofficial, unrecorded and dispersed information. To cope with these limitations, a more intimate approach at the level of neighbourhoods and qualitative approach seem essential in order to integrate and verify the UGI quantitative and city-level assessment. This means considering the multitude of actors present at the very local level, ranging from official service providers e such as state agencies operating at the local level, or private enterprises e to local

3 According to Tashakkori and Teddlie (2003: 711), a multilevel mixed model design “is a design in which QUAL data are collected at one level (e.g., child) and QUAN data are collected at another level (e.g., family) in a concurrent or sequential manner to answer interrelated research questions with multiple approaches (QUAL and QUAL) that are pulled together at the end of the study in the form of ‘global inferences’”. The mixed model design relies on the mixing of QUAN and QUAL approaches at several stages of the study where, conversely, a mixed method design involves the mixing of QUAN and QUAL approaches only at the methods stage of a study.

governments, community-based organisations or organised groups of citizens, and all inhabitants. The proposed application of the UGI consequently relies on a double approach that combines two levels of analysis, through a multilevel mixed model design3 (Fig. 1). The aim of using such a combined approach is to gain insight into the formal and more institutional dimension of urban governance at the municipal level (level 1 of analysis, phase 1), as well as the more informal, local and community-based aspects of governance arrangements (level 2 of analysis, phase 2) for water services. Two main adaptations are performed: the first relates to converting UGI principles and indicators into a research framework that can be analysed qualitatively at the settlement level (phase 2); the second is translation of all adaptable UGI indicators, as well as the questionnaires for the community level of analysis, to water service delivery.4 These amendments are consistent with the suggestion foreseen by UNHabitat to complement the core set of quantitative data with more qualitative surveys, and to develop an explicit thematic focus (UN-HABITAT, 2004a). A comprehensive and integrated urban governance assessment can thus be organised into three different phases. The first two phases relate to the assessment of urban governance at municipal and community levels respectively, which represent the two units of analysis. The third phase concerns the comparison of conclusions stemming from these two levels of analysis. In order to define a common framework for analysis and comparison of the three different phases, 8 sub-criteria have been identified between the 4 UGI core principles, and the 25 UGI indicators. Sub-criteria here

4 UGI indicators' adaptations to the water sector have been done solely for those indicators concerning basic services in general. Originally, we tried to adapt the full twenty-five indicators to the water sector but, for many of them, this change was not possible or, where possible, data was often not available at the municipal level.

438

L. Moretto / Habitat International 49 (2015) 435e444

Table 1 UGI adapted equity indicators for one municipality. No.

Indicator

Data (X)

Formula

Result

Weight

Total

9 10 11 12A 12B 12C 13

Citizens charter for water services (CCS) Percentage of women councillors (WC) Percentage of women in key positions (WK) Percentage of households with water connection (HH wat) Existence of pro-poor water policy (PPC) Is water price cheaper for poor settlements? (WP) Incentives for informal arrangements to improve water access (IA) a. Informal provisions allowed b. Informal provisions with restrictions c. support to informal provisions

No ¼ 0 28.58% n.a. 74.82% Yes ¼ 1 Yes ¼ 1

CC ¼ X WC ¼ X  2/100 WK ¼ X  2/100 HH wat. ¼ 74.82/100 PPC ¼ X WP ¼ X IA ¼ 1 (any one of a, b or c)

0.00 0.57 n.a. 0.75 1.00 1.00 1.00

0.20 0.20 0.10 0.15 0.10 0.10 0.15

0.00 0.114 n.a. 0.112 0.10 0.10 0.15

Yes ¼ 1 No ¼ 0 Yes ¼ 1

Equity sub-index

0.63

proposed are chosen on the basis of the definition of effectiveness, equity, participation and accountability provided by UN-Habitat. For effectiveness and participation, the sub-criteria are mentioned as a “measurable definition” of these principles by UNHabitat (UN-HABITAT, 2004a: 21). For equity and accountability, sub-criteria are deduced from the principles’ explanations provided in the same document. Therefore, data collection, analysis, inferences and meta-inferences are all organised on the basis of the following eight sub-criteria:

UGI principles

Sub-criteria

Effectiveness

Sub-criterion Sub-criterion Sub-criterion Sub-criterion Sub-criterion Sub-criterion Sub-criterion Sub-criterion

Equity Participation Accountability

1: 2: 3: 4: 5: 6: 7: 8:

municipalities was collected in two ways: through direct consultation of official national statistics, which were produced by national state bodies (notably, the Instituto Nacional de Estatistica (INE) and the Consejo Nacional Electoral5), and available municipal budgets and accounts,6 and via interviews with local government and stateowned water company officials, as well as community residents.7 An example of adaptation of UGI principles to water service, and calculation, is provided in Table 1. Quantitative data collected according to the UGI indicators was used to explore local governance in

Subsidiarity of authority, and sufficient resources and autonomy (Institutional) efficiency in delivering public services and responding to civil society concerns and welfare Distributional equity Procedural equity Representative democracy Participative democracy Downward accountability Upward accountability

The three phases of assessment were tested on two municipalities of the Caracas Metropolitan Region in Venezuela, and two specific informal settlements in these municipalities, where innovative water co-production initiatives are taking place. At the core of these initiatives are the “technical water committees” e Mesas Tecnicas de Agua (MTA) e which are community-based organisations, built for the purpose of channelling community participation in the decision-making process and carrying out the physical improvements to service delivery (Allen, 2012; McMillan, Spronk, & Caswell, 2014). Governance arrangements develop around MTAs as a result of agreements, collaboration and cooperation with the water companies, the municipalities and all of the community members (McMillan et al., 2014; Moretto, 2014; Moretto & Allen, 2015).

the municipalities through a “qualitising process”, by which quantitative data is transformed into, and then analysed as, qualitative information8 (Tashakkori & Teddlie, 2010). This process aims at drawing “verbal portraits” (Sandelowski, 2000) or “narrative profiles” (Tashakkori & Teddlie, 1998) concerning the specific target phenomena. In general, “qualitising can be used to extract more information from quantitative data or to confirm interpretations of it” (Sandelowski, 2000: 253). The target phenomena under analysis are represented by the 8 UGI sub-criteria for which as many profiles are drawn. Amongst the main profile types (see Tashakkori & Teddlie, 1998), this analysis relies mainly on a comparative profile,9 where comparison is made against the best score for each indicator. 3.2. Phase 2: community level of analysis

3.1. Phase 1: municipal level of analysis Here the aim is exploring and understanding the extent and the quality of governance at the municipal level in improving water supplies, by applying the UGI indicators. Factual data and information for calculating the UGI indicators in the two selected

5 Generally available through access to the respective websites: http://www.ine. gov.ve and http://www.cne.gov.ve. 6 Municipal budgets and accounts refer to those at the time of the field investigation (2006). To ensure coherence between the municipal data in the budgets and accounts collected from the two municipalities, and other quantitative data available through national statistical databases providing more recent information (such as the percentage of households accessing or voter turnout), all data included in the UGI calculation is from the year 2006 or before. When applicable, our analysis explains essential resulting changes with respect to the situations portrayed up to this date.

The second phase aims at assessing the four basic UGI principles with respect to the informal side, and community dynamics of local governance systems in improving access to urban basic services,

7 In this way e according to UN-HABITAT (2004a) recommendations e the quantitative data has almost always been verified through urban experts and urban policy makers involved in city planning. 8 The qualitising technique applied to the UGI indicators in order to develop subcriteria profiles employed three additional and qualitative data sources. First, UNHabitat concept papers and reports, in order to investigate the significance of the sub-criteria; second, relevant scientific literature about the indicators and subcriteria themes e especially in the context of Venezuela; third, interviews with key actors belonging to the two municipalities, hydrological companies, and academia, in order to integrate the quantitative indicators with further qualitative information. 9 A comparative profile is a narrative description based on the comparison of cases to each other on one or more sets of scores (Tashakkori & Teddlie, 1998).

L. Moretto / Habitat International 49 (2015) 435e444

through a qualitative exploration. Here we examine the degree of involvement and inclusion of low-income communities in these governance arrangements through the technical water committees (MTAs). The unit of analysis in this phase is represented by two informal settlements included in the two previously analysed municipalities although, given their considerable size, only a specific area was chosen in both of them. Semi-structured interviews were the primary mode of qualitative information collection. Different question sets were designed, on the basis of the principle framework of the UGI, for each group of interviewees directly involved in the process of service delivery (the municipality, the water company, the technical water committees, and community residents). The Appendix provides the list of queries asked to all selected interviewees. The number of interviews completed with the community members cannot represent a statistical sample because of the small number. Nevertheless, it is possible to identify a “frequency effect size” (Onwuegbuzie & Teddlie, 2003) related to the rate of the prevalence of certain types of answers given by community members. This operation can also benefit from the option to “binarise” the answers given by residents into a “yes” or “no” answer. Results from interviews with dwellers are, however, always triangulated with other sources of evidence, such as documentation collected at the local level, participant and direct (or systematic) observation, focus groups, and physical artefacts. 3.3. Phase 3: meta-inferences from the two units of analysis This methodological approach leads to a comparison between the inferences resulting from the previous two phases. The analysis in this phase aims at understanding, in an integrated way, the two facets of urban governance for water delivery. The integration of quantitative (municipal level) and qualitative (community level) approaches may provide three different outcomes, as a comparison of results can be convergent, divergent or complementary (Erzberger & Kelle, 2003). Meta-inferences from the two levels and units of analysis are worked out according to a narrative integration and comparison of the “qualitised” inferences belonging to phase 1 and the qualitative results belonging to phase 2 of the study. This stage of analysis rests primarily on a common theoretical framework provided by the Urban Governance index principle framework. Secondly, comparisons are made following two different models. A “mutual validation model”, which searches “for convergent findings as the most important purpose”, and a “complementary model”, which “draws our attention to the fact that different methods may relate to different aspects of the domain in question because certain methods have been developed within different theoretical traditions and thus serve the investigation of different phenomena” (Erzberger & Kelle, 2003: 484). 4. Urban governance at the municipal and settlement levels Two examples of meta-inferences, based on the results stemming from phase 1 and 2 of assessments are presented here. 4.1. Sub-criterion 1: subsidiarity of authority and sufficient resources and autonomy Despite a comprehensive normative framework aimed at completing and consolidating the decentralisation process initiated in Venezuela in 1989, and involving water and sanitation services (Asamblea Nacional de la Republica Bolivariana de Venezuela, 2001; 2007), investigation at the municipal level brings to the forefront a situation in which local governments have an unclear level of power and limited financial resources in the operation of basic services, including water. In practice, the state-owned water company

439

operating in the two selected municipalities continues to be the only official institution supplying drinking water, working directly with low-income communities and organising the technical water committees. This tendency is confirmed by the qualitative analysis run in the two low-income settlements. Interviews with MTA leaders, municipal public servants, and water company officials confirm that, in one settlement, responsibility for the realisation and operation of the piped network has been principally shared between the community, the MTA and the public water company while, in the other, community members and MTA leaders have mainly relied on selfmade solutions to their water shortages. The incomplete decentralisation of power and financial resources available to municipalities, and the concurrent delegation of responsibilities and economic means to communities, therefore makes the current local governance arrangements for improving drinking water provisions heavily dependent on central government policies (Moretto, 2014; Moretto & Allen, 2015). Results stemming from analysis at both the municipal and community levels are thus, at the same time, convergent and complementary in showing that responsibilities for water service operation appear entrusted to the technical water committees, rather than the municipal offices charged with it. But the effectiveness of local governance arrangements is also affected by the specific water access conditions at the settlement level, so UGI indicators cannot provide information applicable to entire settlements within the municipality. The existence, for instance, of somewhat illegal connections to the water mains in one of the studied settlements, although informal, unreliable, and not accessible to all dwellers, together with a long-standing fuzziness in the previous administrative zoning of the area, have represented two major limitations in the effectiveness of local arrangements, and the ensuing delivery of an effective service. This has resulted in all three actors e the community, the municipality and the water company e shifting responsibility onto each other. 4.2. Sub-criterion 3: distributional equity Although UGI equity indicators portray a situation in which drinking water provisions through piped connections are widespread in the studied areas, and their accessibility is secured by propoor policies, local access conditions may indicate severe shortcomings in the different low-income communities. Divergent results between the two phases of analysis are clear if we compare the relatively high percentage of households connected to a piped water network in the two municipalities (74.82% and 92.21%; INE, 2001a, b) with the very different water access conditions, not only between the two settlements, but also within the same community. Household connections in one of the two settlements were considered well-performing by all of the local actors, as water was available one day out of two for all of the residents. Nevertheless, many households received water with varied pressure and frequency according to their relative elevation and distance from the main pipes, while others were almost completely lacking the service because of the new illegal connections by neighbourhood settlements in the same network of mains. In the other area, the inequity in accessing water was even worse. Some areas were provided with quite a good service, night and day; some others accessed water through illegal connections just 6e7 h a day, in small quantities and with low pressure; others were totally lacking in any form of access to water. Results from the qualitative investigation at the community level complement the optimistic figures stemming from the 2001 and 2011 Venezuelan National Surveys, which do not report problems related to the quality of the service and, at the same time, confirm that in low-income communities, an extremely high percentage of the population does not have continuous access to water.

440

L. Moretto / Habitat International 49 (2015) 435e444

On the contrary, convergent results are found with respect to the affordability of the service. For instance, the existence of a propoor pricing policy (UGI indicator 12B) allows residents to reduce the price to service access. From one side, the new organic law on water and sanitation (Asamblea Nacional de la Republica Bolivariana de Venezuela, 2001; 2007) provides the legal basis for an efficient and cost-effective water service with affordable prices e especially for low-income residents e aimed at equitable access to water, together with the sharing of responsibilities for the water services. From the other, the analysis at the community level shows the concrete consequences of this policy. First, construction and connection costs for low-income residents are dramatically reduced as they are shared amongst the water company, the community and e to a limited extent e the municipality. Second, when applicable, citizens are charged for water with a minimum social monthly tariff. Third, willingness to pay is high as citizens see water payment as a way of being acknowledged in the settlement, of getting an institutional response to their water shortage issues, and of receiving assistance in tackling the network inefficiencies. 5. Experience with the assessment tool The application of the UGI to the arrangements developed in Venezuela for improving water supplies has been carried out through two major adaptations. The most important revision concerned an experimental review of the UGI principles and indicators in order to elaborate a set of queries for assessing how the urban governance arrangements, aimed at improving access to the water supply for the urban poor, work on the ground. This is in line with a move from a topedown to a bottomeup approach in governance evaluation, but at a very local level where “citizens assess how well [their] needs and concerns are being met” (Wilde, 2011: 41) according to their own principles and priorities, and not only with respect to the criteria stated elsewhere10. By translating the UGI's principle framework into a more qualitative and settlement-based investigation, the Index has proved capable of providing a clear and comprehensive structure to shape analysis at the community level, in two main regards. First, this operation helped to configure the data collection for exploring the more informal side and community aspects of local governance systems, as shown in Appendix 1. Second e and most importantly e it allowed for mixing the outcomes of the two units of analysis, on the basis of a common theoretical and methodological framework. Further, the UGI has also been reworked according to the specific governance policies, mechanisms and practices for drinking water supplies. This adaptation has included both the municipal and community levels of analysis, allowing comparison of results from these different perspectives. At the municipal level, the focus on water provisions allowed appreciation of the extent of decentralisation in the water sector and the principal relationships amongst the key actors involved in the service delivery process. At the community level, this focus permitted a look inside the current and innovative local governance arrangements aimed at improving water supplies within two specific settlements. Given the

10 Wilde (2011) considers the Urban Governance Index as a “third generation” assessment tool, defining this category as explicitly addressing a bottom up process in governance assessments, which is supposed to be inclusive, transparent, and more democratic. But our analysis of the UGI has not focused on the process of governance evaluation (data collection was, in effect, not participatory as UGI calculation was performed in the context of a research study) but rather, on the structure and principle framework of the evaluation tool. It is in this sense that we believe an explicit focus on the settlement/community level of governance could further contribute to making governance assessments tools, such as the UGI, more inclusive, transparent, democratic and participatory.

importance that the UGI attributes to a “principles framework” that emphasises “the quality of relationships and processes between key stakeholders at the local level” (UN-HABITAT, 2004a: 18) e rather than a framework based just on the quantity and performance of local government service delivery e it has been possible to investigate and compare governance mechanisms built around the water delivery process with respect to their dynamics and processes at both the municipal and community levels. The Index has thus proved to be a flexible, handy, and helpful instrument that is adaptable and applicable to diverse contexts, levels and approaches. The UGI in itself, in the way it is presented by UN-Habitat, cannot be considered an accurate tool to investigate the dual nature of urban governance since the most local level of governance, which is represented by the settlement and its inhabiting community, is excluded from UGI indicators. This is particularly striking when we see to what extent access to basic services, and inclusion in local decision-making, can vary, not only among settlements within the same municipality, but also within the same settlement, as shown in our analysis. 6. Policy implications and inputs for future research Regardless of the correlation between inferences at the municipal and community levels (divergent, convergent or complementary), it is clear that the Urban Governance Index appears to at least partially fail in comprehensively appreciating the local dynamics of urban governance for the improvement of services. Three main considerations arise for current and future application of the UGI. 6.1. Settlement unit of analysis Complementing the UGI with an assessment at a more local and community-centred level constitutes not only a possibility noted by the Index (UN-HABITAT, 2004a), but also a necessity. Evidence from this research reveals the possible gap between improved service provisions and the governance framework in which these improvements take place. On one hand, most improvements in access to water supplies for the urban poor in the Caracas Metropolitan Region are directly linked to the mobilisation of communities and their commitment to applying pressure on local institutions, rather than to municipal initiatives. In assessing the quality of urban governance processes and relationships, the extent of the whole community's inclusion in the local agreements, and the quality of the relationships among local actors on the ground e including community residents e cannot be excluded. On the other hand, improvements in the delivery of drinking water can be linked to different dynamics than those directly related to the performance of local administration and the decentralisation of levels involved in service delivery. First, better water access for low-income communities does not always depend on a strong representative democracy at the municipal level, on the establishment of accountability mechanisms within the local administration, or on the presence of a pro-poor pricing policy. The UGI's focus on local governments, as the unit ensuring effective local governance, can result in a limited understanding of why some small-scale, informal, and community-based experiences work. Second, nationally-based public institutions (such as the central governments or public utilities) may also substantially contribute to improved water provisions. UGI assessment is thus likely to exclude those local governance practices which result in relevant improvements in basic service provisions, and involve actors belonging to the public, private (in its diverse variants) and community sectors, but, at the same time, place less emphasis on the municipal level e as in the case of water services in Venezuela. Further development of

L. Moretto / Habitat International 49 (2015) 435e444

the UGI should more systematically promote the integration of urban governance assessments at the local level of analysis, in order to thoroughly investigate how governance works in practice: primarily, at a community level and, subsequently, in relation to institutional levels other than local governments. 6.2. Qualitative methods The integration of qualitative approaches into the UGI could face severe limitations. Primarily, qualitative surveys are generally built according to unique contexts and needs and, hence, are rarely transferable to other circumstances. It would be difficult to meet the “universality” criterion established by UN-Habitat in elaborating the UGI, but also, in some cases, the criteria of “relevance” and “ease of collection”. Next, comparisons between quantitative and qualitative data should be run on one of these two approaches (either quantitative of qualitative). Quantifying qualitative information and aggregating it into more synthetic indexes or sub-indexes might be in contrast with the very nature and scope of a qualitative approach, which is to provide the most detailed portrayal as possible. In contrast, qualitising quantitative information poses the risk of blurring the UGI objective to “enable cities to objectively measure the quality of local governance” (Narang, 2005a, 2005b: 1). Nevertheless, a qualitative approach to exploring local governance arrangements at the community level has proved to be extremely productive for two main reasons. On one hand, it reveals aspects of the local governance arrangements that the simple UGI indicators might not disclose, by extracting more information from the quantitative data,11 while at the same time double-checking findings from the calculation of the Urban Governance Indicators. On the other hand, it provides an appropriate methodological framework e often the only method e to explore informal and mutual governance agreements that are, in reality, often excluded from a more quantitative methodological approach that looks for formal and easily quantifiable information. The narrative profiles previously presented (see “Urban governance at the municipal and settlement levels”) actually fit well with some of the shifts identin (2011) in governance assessments over the last two fied by Hyde decades, notably “from numerical indicators to narrative trajectories”, as the objects of governance evaluations are always sitespecific, and from an “exclusive reliance on quantitative data … [to] a greater readiness to adopt and use qualitative methods of n, 2011: 17). analysis” (Hyde 6.3. Normative framework The results stemming from analysis at the municipal/settlement level raise a more general and conceptual question about the significance and adequacy of the UGI principle framework to evaluate urban governance. Although “all [governance] assessments have a normative bias otherwise it is not possible to judge what is good or bad or if something is improving or deteriorating” (Wilde, 2011: 36), the analysis presented here has shown how citizens’ access to better basic services and effective decision-making has often been reached without, or outside of, governance systems that fulfilled the principles of effectiveness, equity, participation and accountability. Two major outcomes from this research confirm this assumption. First, a gap exists between effective improvements in water delivery and the high score of urban governance indicators. In our case studies, progresses in service provision were linked much more with

11 For instance, this has been the case with the analysis of the decentralisation process in Venezuela to understand the reasons and significance of most of the UGI effectiveness indicators.

441

the local capacity (especially from the water community-based associations and the public water company) to elaborate rapid, operational and appropriate solutions to water shortage e often in informal and unorthodox ways e rather than to high UGI scores on sub-indexes. Incomplete service decentralisation, and the resulting low scores on UGI indicators of effectiveness, or the absence of specific municipal bodies for local administration transparency, responsiveness, and integrity to ensure downward accountability in urban governance, did not prevent locally-developed and community-based governance arrangements from making structural improvements to water access. On the other hand, good equity indicator scores (related, for instance, to the existence of a pro-poor pricing policy, cheaper water in poor settlements, or incentives for informal agreements to improve water) did not stop community residents from experiencing large differences in the quality, frequency, and reliability of their water access, resulting in sociospatial micro-fragmentations within the settlement. Second, a gap also exists between upgrading material water delivery and the UGI assumptions behind the principle framework. For example, effective water improvements were actually reached in cases where community member participation in the decisionmaking process e and thus in local democracy e was quite limited. These include, for instance, cases where community members took part in the realisation of the network but were generally excluded from time-consuming meetings, or where water committee leaders engaged directly with local government representatives to get a quick response to water shortage issues rather than spending time undertaking official procedures to assure formal upward accountability. These cases raise some general questions about the capacity of the principles of effectiveness, equity, participation and accountability to really correlate with the way in which governance works in practice. Here we discuss two examples. Looking for instance at the participatory dimension of governance, it seems appropriate to wonder whether it is legitimate to link it to democracy. We know that the number of civic associations cannot be a measure of participative democracy when the requirement for citizens to be grouped in associations that can engage the state could risk excluding, at the very local level, the less well networked and organised groups, the powerless voices within social groups, and also the individuals outside these groups (Batley & Mcloughlin, 2010; Lombard, 2013; Maloutas & Malouta, 2004; Verschuere, Brandsen, & Pestoff, 2012). We also know that in some specific political contexts, such as in Venezuela, the autonomy and independence of community-based organisations CBOs cannot be taken for granted (Lopez Maya & Lander, 2011) as “participation does not necessarily lead to anything we would recognise as democratisation” (Smilde, 2011: 17), and service co-production runs the risk of not really challenging conventional forms of service delivery through grassroots organisations (Mitlin, 2008). Participation is hence certainly broader than what is proposed in the UGI indicators. So, possibly, as suggested by some scholars (Hickey, 2010; UNHABITAT and Al-Madinah Al-Munawarah, 2010; Wilde, 2011), the participatory and bottomeup dimension of governance should probably include a “right-based approach”, with a much larger scope than representative and participatory democracy.12 Turning to the example of accountability, evidence from research stresses how the boundaries between the spheres of the

12 Referring to a case in San Salvador during an Expert Group Meeting on Localising the Habitat Agenda Indicators: New paradigms in developing indicators, Yves Cabannes focused, for instance, on the “right to participate”, while underlining that this includes “the right to request and receive information, the right to consult and propose, the right to participate in the decision making process, the right to comanagement, the oversight right and the right to denounce” (UN-HABITAT and Al-Madinah Al-Munawarah, 2010: 8).

442

L. Moretto / Habitat International 49 (2015) 435e444

local stakeholders taking part in the “tiers of governance” (for public services) are blurred and imprecise (Benequista & Gaventa, 2011; Bovaird, 2007; Oia, 2011). Research shows that mutual agreements between the water company and the water committees in Venezuela mainly developed outside formal procedures, codified rules and official administrative structures and bodies, mixing the roles of these two actors in the service delivery process. The question thus arises of how credibility and accountability could be ensured through relationships between the state and citizens that are “undefined, informal and renegotiated almost continuously” (Joshi & Moore, 2004), rather than through clear and enforceable contracts (Batley, 2006; McDonald & Ruiters, 2012; Ostrom, 1996). Put another way, can formal, explicit, and direct accountability as defined by UN-Habitat encompass the increasingly undefined, informal and blurred governance relationships among local actors in service provision? The divergence of reality from normative prescriptions when governance operates on the ground is not a new concern (Joshi & Moore, 2004; Tunstall, 2001) and, hopefully, “today governance relies less on normative blueprints and more on practical

Community members Effectiveness Effective delivery of public services

Water board leaders

n, 2011: 19). Governance assessments e experimentation” (Hyde including the Urban Governance Index e should, as a consequence, critically question the relevance of a normative and ideological framework in understanding how governance works in practice. They should also systematically integrate a more qualitative, community-level, and context-specific dimension in their evaluation architecture, to include the very local dimension of governance on the ground. Acknowledgements Thanks to the Italian Ministry of Education and Research for funding my doctoral studies, as the content of this article is part of my PhD thesis that was recently published by PIE Peter Lang in the  libre de Bruxelles, my form of a book. Thanks also to the Universite employer, for granting me the time to revise my research. Appendix. Semi-structured interview framework (phase 2).

Water board focus group

Municipal public servants

- Do you think that the - Up to now, how do you - What are the results of and - What are the results of and changes in MTA changes in MTA activities in the think that the MTA has MTA has been activities in the community? contributed to improving important in improving - What are the limitations, community? water access in the your access to water? difficulties and problems in the community? - How do you think the MTA MTA activities? could improve the current water supply for the whole community? does the Responsiveness - Did any of the MTA - How does the MTA become - What could the MTA do to - How municipality become improve the water supply for aware of community water leaders ask you about to civil aware of community the community? related problems? your water shortage? society water related problems concerns and MTA activities? Equity Access to basic - Up to now, how has - What is the purpose of the - What is the purpose of the MTA? - What is the municipalʼs MTA? aim with regards to water services your access to water supply for the (selected) improved? community? - How do you think that the MTA could improve your current access? does the - What do you think - Was there a meeting to - How do you think the water - How Equity in municipality involve the payment should be organised ? speak about water about water payment? decision - How did you speak about the community/MTA in the payment? - Do you pay for water? making decision-making process? much does the water supply payment with the - Did you agree to pay for - How - What is the municipal community and with community pay for water? water? position with regards to - Did you share your - Did the MTA ask the Hidrocapital? water subsidies? community about the opinion with an MTA water payment? leader? Participation many people - What are the relationships with - What are the relationships Participative - Do you know the MTA? - How democracy - How did you hear about participate at the MTA with the MTA? the community, the other the MTA? - What are the relationships community based association, meetings? - Have you ever spoken - What are the relationships the Municipality and with Hidrocapital? about your water Hidrocapital? with the Municipality and related problems with Hidrocapital? an MTA leader? - Is there a ʻwater community - Did you collaborate at councilʼ? the pipeline construction? Representative - Were you aware of the - How were the MTA leaders - How were the MTA leaders - How are the MTA leaders elected? elected? chosen? democracy meeting in which the MTA leaders were elected?

State-owned water company officials - What are the results of and changes in MTA activities in the community?

- How does Hidrocapital become aware of community water related problems and MTA activities? - What is Hidrocapitalʼs purpose with regards to water supply for the (selected) community?

How does Hidrocapital involve the community/ MTA in the decisionmaking process? - How is the tariff structure organised?

- What are the relationships with the MTA? - What are the relationships with the municipality?

- How are the MTA leaders chosen?

L. Moretto / Habitat International 49 (2015) 435e444

443

(continued ) Community members

Water board leaders

Water board focus group

Accountability Transparency

Municipal public servants

- Do you know the - How does the MTA inform - How does the MTA inform the - What are the municipal responsibilities with other stakeholders about the the community about purpose of the MTA? regards to water supply? changes and decision about - Do you know the meetingsʼ with the - How does the water supply? municipality and outcomes of the MTA Municipality inform the Hidrocapital? meeting? community/MTA about its responsibilities? Integrity - Is there an MTA written - What and how are the (UGI quantitative indicators) regulation? relationships with the other community-based committees (as the CTU)? - How often do you meet Responsiveness - Did any of the MTA - How does the MTA become the MTA leaders? aware of community water leaders ask you about - Is there an office where related problems? your water shortage? people can complain? Willingness to responsibilities - Are you willing to be take and actively involved in the share pipeline construction and in the MTA meetings? - Did you collaborate in the pipeline construction?

References Allen, A. (2012). Water provision for and by the peri-urban poor: public-community partnerships or citizens co-production? In I. Vojnovic (Ed.), Sustainability: A global urban context (pp. 309e340) East Lansing MI: Michigan State University Press. Allen, A., Davila, J., & Hofman, P. (2006). The peri-urban water poor: citizens or consumers? Environment and Urbanization, 18(2), 333e351. Amis, P., Batley, R., Beall, J., Devas, N., Grant, U., Kanji, N., et al. (2001). Urban governance and poverty: Lessons from a study of ten cities in the south. Birmingham: The School of the Public Policy, University of Birmingham. Asamblea Nacional de la Republica Bolivariana de Venezuela. (2001). Ley Organica para la Prestacion de los Servicios de Agua Potable y Saneamiento. Gaceta Oficial Extraordinaria No. 5568, 31st of December 2001. Asamblea Nacional de la Republica Bolivariana de Venezuela. (2007). Ley de Reforma nica para la Prestacio n de los Servicios de Agua Potable y Parcial de la Ley Orga Saneamiento. Gaceta Oficial No. 38763, 6th of December 2007. Bakker, K. (2010). Privatizing water: Governance failure and the World's urban water crisis. Ithaca, NY: Cornell University Press. Batley, R. (2006). Engaged or divorced? Cross-service findings on government relations with non-state service-providers (Vol. 26, pp. 241e251). Public Administration and Development. No. 3. Batley, R., & Mcloughlin, C. (2010). Engagement with non-state service providers in fragile states: reconciling state-building and service delivery. Development Policy Review, 28(2), 131e154. Benequista, N., & Gaventa, J. (2011). Blurring the boundaries. Citizen action across states and societies. Brighton: Development Centre on Citizenship, Participation and Accountability. Booth, D. (2011). Towards a theory of local governance and public goods provision (Vol. 42, pp. 11e21). IDS Bulletin. no. 2. Bovaird, T. (2007). Beyond engagement and participation: User and community coproduction of public services (Vol. 67, pp. 846e860). Public Administration Review. No. 5. Burgess, R., Carmona, M., & Kolstee, T. (Eds.). (1997). The challenge of sustainable cities. London: Zed Books Ltd. Erzberger, C., & Kelle, U. (2003). Making inferences in mixed methods: the rules of integration. In A. Tashakkori, & C. Teddlie (Eds.), Handbook of mixed methods in social and behavioural research. Thousands Oaks, California: SAGE Publications. Gaventa, J., & Barrett, G. (2010). So what difference does it make? Mapping the outcomes of citizen engagement. IDS Working Paper 347. Halfani, M., McCarney, P., & Rodriguez, A. (1995). Towards an understanding of governance. In R. Stren, & J. Bell (Eds.), Urban research in the developing world (Vol. 4). Perspective on the City (Toronto: Centre for Urban and Community Studies, University of Toronto), 91e14. Harpham, T., & Boateng, K. (1997). Urban governance in relation to the operation of urban services in developing Countries. Habitat International, 21(1), 65e77. Hickey, S. (2010). The government of chronic poverty: from exclusion to citizenship? The Journal of Development Studies, 46(7), 1139e1155. n, G.. (2011). Making the state responsive: rethinking governance theory and Hyde n, & J. Samuel (Eds.), Making the state responsive: Experience practice. In G.. Hyde with democratic governance assessments (pp. 5e28). United Nations Development programme.

State-owned water company officials - What are Hidrocapital responsibilities with regard to water supply? - How does Hidrocapital inform the community/ MTA about its responsibilities? (UGI quantitative indicators)

- How often do you meet the MTA leaders? - Is there an office where people can complain?

n, G., & Samuel, J. (Eds.). (2011). Making the state responsive: Experience with Hyde democratic governance assessments. United Nations Development programme. INE e Instituto Nacional de Estadistica. (2001a). Censo 2001. www.ine.gov.ve. Assessing Urban Governance 234. INE e Instituto Nacional de Estadistica. (2001b). XIV Cenco National de Poblacion y Vivienda. www.ine.gov.ve. Jessop, B. (2002). Liberalism, neoliberalism, and urban governance: A state-theoretical perspective’, in antipode (pp. 452e472). Blackwell Publishers. Joshi, A., & Moore, M. (2004). Institutionalised co-production: unorthodox public service delivery in challenging environments. The Journal of Development Studies, 40(4), 31e49. autier, F., & Mastruzzi, M. (2004). Governance and the city: An Kaufmann, D., Le empirical exploration into global determinants of urban performance. World Bank Institute (Discussion Paper). Lombard, M. (2013). Citizen participation in urban governance in the context of democratization: evidence from low-income neighbourhoods in Mexico. International Journal of Urban and Regional Research, 37(1), 135e150. Lopez Maya, M., & Lander, L. E. (2011). Participatory democracy in Venezuela: origins, ideas, and implementation. In D. Smilde, & D. Hellinger (Eds.), Venezuela's vez (pp. bolivarian democracy: Participation, politics, and culture under Cha 58e79). Duke University Press. Maloutas, T., & Malouta, M. P. (2004). The glass managerie of urban governance and social cohesion: concepts and stakes/concepts as stakes. International Journal of Urban and Regional Research, 28(2), 449e465 (Oxford). McDonald, D., & Ruiters, G. (2012). Alternatives to privatization: Public options for essential services in the global South. South Africa: Routledge e HSRC e LeftWord Books. McMillan, R., Spronk, S., & Caswell, C. (2014). Popular participation, equity, and coproduction of water and sanitation services in Caracas, Venezuela. Water International, 39(2), 201e215. Miegville, S., Ramirez, T., Ricard, M., Ropert, A., & Wagner, A.-G. (2003). La gouvernance urbaine : Un concept polys emique op eratoire pour les villes des pays en me e conomiques du de veloppement, mimeo. d eveloppement?, S eminaire. Proble Mitlin, D. (2008). With and beyond the state e co-production as a route to political influence, power and transformation for grassroots organizations. Environment and Urbanization, 20(2), 339e360. Moretto, L. (2007). Urban governance and multilateral aid organizations: The case of informal water supply systems (Vol. 2, pp. 345e370). The Review of International Organizations. no. 4. Moretto, L. (2014). Assessing Urban Governance: The case of water service coproduction in Venezuela. Brussels: PIE Petr Lang. Moretto, L., & Allen, A. (2015). Co-producing water services in peri-urban caracas: political democratisation without administrative decentralisation? In J. Bogaert, & J. M. Halleux (Eds.), 2015, Territoires p eriurbains : d eveloppement, enjeux et ge, Belgium: Presses Agronomiques perspectives dans les pays du Sud, Edition Lie de Gembloux. Narang, S. (2005a). Disaggregating governance indicators: Why local governance is important and how it can be measured. paper presented at the Technical Workshop on Governance Indictors, April 20e22, New Delhi, India. Narang, S. (2005b). Disaggregating governance indicators: Why local governance is important and how it can be measured. paper presented at the Technical Workshop on Governance Indictors, April 20e22, New Delhi, India.

444

L. Moretto / Habitat International 49 (2015) 435e444

Nickson, A., & Franceys, R. (2003). Tapping the market. The Challenge of institutional reform in the urban sector. Houndmills and New York: Palgrave. Obeng-Odoom, F. (2012). On the origin, meaning, and evaluation of urban governance. Norsk Geografisk Tidsskrift-Norwegian Journal of Geography, 66(4), 204e212. Oia, I. (2011). Squaring governance assessments with the accountability agenda. In n, & J. Samuel (Eds.), 2011, making the state Responsive: Experience with G.. Hyde democratic governance assessments (pp. 63e82). United Nations Development programme. Olivier de Sardan, J. P. (2009). The eight modes of local governance in West Africa. Working Paper no. 4, APPP. Onwuegbuzie, A. J., & Teddlie, C. (2003). A framework for analysing data in mixed methods research. In A. Tashakkori, & C. Teddlie (Eds.), Handbook of mixed methods in social and behavioural research. Thousands Oaks, California: SAGE Publications. Ostrom, E. (1996). Crossing the great Divide: Coproduction, Synergy, and development. World Development, 24(6), 1073e1087. Paproski, P. (1993). Urban governance systemseanother unanalysed abstraction? London: DPU NEWS. Issue no. 28, February. Richard, B., & Larbi, G. (2004). The changing role of government. The reform of public services in developing countries. Houndmills and New York: Palgrave. Richard, B., & Moran, D. (2004). Literature review of non-state provision of basic services. Birmingham: International Development Department, School of Public Policy, The University of Birmingham. Sandelowski, M. (2000). Combining qualitative and quantitative sampling, data collection and analysis techniques in mixed-methods studies. Research in Nursing and Health, 23, 246e255. Smilde, D. (2011). Introduction: participation, politics, and culture e emerging fragments of Venezuela's Bolivarian democracy. In D. Smilde, & D. Hellinger (Eds.), Venezuela's Bolivarian Democracy: Participation, politics, and culture under vez (pp. 1e27). Duke University Press. Cha Stewart, K. (2006). Designing good urban governance indicators: the importance of citizens participation and its evaluation in greater Vancouver. Cities, 23(3), 196e204. Stren, R. (2012). Donor assistance and urban service delivery in Africa. United Nations

University, World Institute for Development Economics Research. Working Paper No. 2012/49. Swingedouw, E. (2005). Governance innovation and the citizen: the janus face of governance-beyond-the-state. Urban Studies, 42(11), 1991e2006. Tashakkori, A., & Teddlie, C. (1998). Mixed methodology: Combining qualitative and quantitative approaches. Thousands Oaks, California: SAGE Publications. Tashakkori, A., & Teddlie, C. (2003). Handbook of mixed methods in social and behavioural research. Thousands Oaks, California: SAGE Publications. Tashakkori, A., & Teddlie, C. (2010). Handbook of mixed methods in social and behavioural research (2nd ed.). Thousands Oaks, California: SAGE Publications. Tunstall, R. (2001). Devolution and user participation in public services: how they work and what they do. Urban Studies, 38(13), 2495e2514. UN-HABITAT. (2002). The global campaign for good urban governance (2nd ed.) Concept Paper www.unhabitat.org. UN-HABITAT. (2004a). Urban governance Index: Conceptual foundation and field Test report (Nairobi, Kenya). UN-HABITAT. (2004b). Urban indicators Guidelines: Monitoring the habitat agenda and the millennium development goals. www.unhabitat.org/governance. UN-HABITAT. (2004c). Guidelines for conducting the urban governance Index; addendum to the urban governance index conceptual foundation and field test report. UN-HABITAT and Al-Madinah Al-Munawarah. (2010). Expert group meeting, localising the habitat agenda indicators: New paradigms in developing indicators. Final Report, 8e10 February 2010. UNDP. (2007). Governance indicators: A users' guide (2nd ed.). Verschuere, B., Brandsen, T., & Pestoff, V. (2012). Co-production: the state of the art in research and the future agenda. Voluntas, 23(4), 1083e1101. Wild, L., Chambers, V., King, M., & Harris, D. (2012). Common constraints and incentive problems in service delivery. Overseas Development Institute. Working Paper 351. n, & Wilde, A. (2011). The democratization of governance assessments. In G.. Hyde J. Samuel (Eds.), Making the state responsive: Experience with democratic governance assessments (pp. 29e48). United Nations Development programme. Wilde, A., Shipra, N., Laberge, M., & Moretto, L. (2009). A users' guide to measuring local governance. UNDP, Oslo Governance Centre.