Designing public-private partnership for sustainable urban water and sanitation services : a case study of Addis Ababa, Ethiopia

(1)

1 UNIVERSITEIT VAN AMSTERDAM

(2)

Abstract

Public-private partnerships (PPPs) have during the last decades consolidated as an instrument for reaching development objectives in many sectors across the world. A range of donors and policy makers argue that PPPs contribute to economic growth through job creation, technology transfer, capacity development, innovation, and that they stimulate investment in local economies with financial injections. Critics argue that PPPs might result in services at higher cost, of poorer quality and limited quantity for marginalised groups. Due to divergent experiences across developed and developing countries, controversy around the suitability of PPPs for public service provision and deviating theoretical understandings of PPPs, this thesis aims to investigate if the theoretical fragmentation of the concept has implications for a PPP in practice. Through a case study of a PPP in the WaSH sector in Addis Ababa it addresses the question: How can public-private partnerships be designed in order to contribute to sustainable urban WaSH services? Through combining an extensive literature review with an instrumental analysis of one WaSH PPP, the study adds to the discussion about PPP design that can be further tested in future research and application of PPPs. The retroductive research approach applies qualitative methods including document analysis, qualitative content analysis and semi-structured interviews. The study finds that i) there is a contrasting understanding about the use and format of PPPs, ii) there is a divergent understanding of WaSH as either a public or a private good, iii) the contracts, project plans and budgets are confidential which makes informed assessment of the effectiveness and long-term sustainability of PPPs impossible, and iv) partial PPPs that exclude important interventions on sanitation might negatively affect communities further downstream. Hence, the study suggests that PPPs aiming at development goals and financed by donor or recipient government funds need to i) ensure shared understanding of partnership practice, ii) provide clear provisions on how to manage public goods within PPPs, iii) apply transparent contracts and appropriate fund use, and iv) integrate WaSH services to ensure long-term sustainability. This study argues that a mutual understanding of the partnership and of WaSH is a crucial factor for providing sustainable WaSH services. With this, the study exposes a need for further research about a potential contradiction between profitability and sustainability of PPPs.

Key words: public-private partnerships/PPPs, urban water and sanitation supply/WaSH, partnership design, Sustainable Development, Addis Ababa.

(3)

Acknowledgements

This thesis is the result of over a year’s work consisting of heavy thinking sessions, hard but fruitful discussions, severe doubts, and complete joy when reaching research breakthroughs. It is also the result from a challenging but rewarding time of data collection in Addis Ababa, Ethiopia during the fall of 2015.

I want to extend my greatest gratitude to my supervisor Prof. Dr. Joyeeta Gupta for guiding me throughout the process, for insightful comments and suggestions, and for challenging and encouraging me in carrying out my own research and present my work to an expanding audience. To Dr. Courtney Vegelin, who kindly agreed to be the second reader of this thesis, and thereby to thoroughly assess my research contribution. To Ato Berhanu Solomon Genet, with his incredibly engaging and inviting involvement shared his valuable time and network in the data collection phase of the research, and kindly accepting the role of local supervisor. To Mr. Daniel Truneh, who was my first, and so crucial, point of contact to the case study location and partners. To Ato Abebe, who helped me attain all necessary data of the Addis Ababa Water and Sewerage Authority about the current water and sanitation situation in Addis. And, to Mr. Johnson Nkem, who so kindly invited me to the UNECA for fruitful discussions and to share his expertise, knowledge and network, and also highly motivated me to continue with my initial research idea when in doubt.

I want to thank all the respondents and participants for agreeing to be part of my research. Without their positive approach to participation, I would not have succeeded.

Finally, I want to acknowledge the people around me who made the stressful times much more pleasant. Tom, my true side-kick, and Els with Dolly and Bucket (Eva!) in Addis. The whole ‘social study crew’ in Amsterdam, my family and friends back home in Sweden and Lena for your constant friendship, discussion and advice, coffees, laughs and cries. And Dr. Garganoa, wherever we were (perhaps more often apart than together during this time), you have been my biggest and most important support in this endeavour.

(4)

List of figures and tables

Figure 1. Facts about current WaSH challenges in the world ... 9

Figure 2. Testimony from the people of Addis Ababa ... 10

Figure 3. Three overlapping domains of reality of Critical Realism ... 16

Figure 4. Conceptual framework for instrumental analysis: concepts and relations between concepts (revised from Young et al., 2005) ... 19

Figure 5. Phases of research ... 23

Figure 6. Main distinctions between the public and the private sector ... 34

Figure 7. Total investment and number of projects with private participation in key infrastructure sectors* in Sub-Saharan Africa, 1990-2015 (US$ billion) ... 35

Figure 8. Total investment and number of projects with private participation in water and sewerage infrastructure in Sub-Saharan Africa, 1990-2015 (US$ billion) ... 36

Figure 9. Total number of projects in water and sanitation by segment in Sub-Saharan Africa 1991-2011 ... 36

Figure 10. Total investment and number of projects in water and sewerage infrastructure in the world, 1990-2015* ... 37

Figure 11. Countries recognizing human right to water and sanitation in constitution or law (n=39) ... 39

Figure 12. The spectrum of public-private partnerships ... 41

Figure 13. Typology of contractual agreements of PPPs ... 47

Figure 14. The four C's of strategic alliance research ... 49

Figure 15. Means of influence and expectations of SDGs Goal #17, the ‘Aid and Trade’ policy of The Netherlands and the ‘GTP2’ plan of Ethiopia on private sector involvement/PPPs ... 57

Figure 16. Definition of S2T&B on the PPP spectrum ... 66

Figure 17. Location of Ethiopia, the Awash Basin and Addis Ababa ... 73

Figure 18. Map of catchment points and approximate elevation levels for Addis Ababa water supply ... 74

Figure 19. Water situation map of Addis Ababa with figures of water service supply to the different sub-cities ... 76

Figure 20. Open water source in urban Addis Ababa, Old Airport Area, November 2015 ... 78

Figure 21. Overview of anticipated impacts of S2T&B ... 82

Figure 22. Overview of PPP impacts on the WaSH system chain in Addis Ababa ... 83

Table 1. Overview of key words and concepts used for literature search ... 17

Table 2. Operationalisation of concepts and their respective indicators ... 21

Table 3. Overview of documents ... 25

Table 4. Overview of types of respondents and respondent categories ... 26

Table 5. Coding scheme ... 27

Table 6. Types and examples of different goods ... 40

Table 7. Overview of improved urban and rural access to water and sanitation services in Ethiopia 1996-2015 ... 53

Table 8. Overiew of compatibility of means of influence and expectations of institutions ... 62

Table 9. Overview of the S2T&B partners, their capacities and financial input ... 65

Table 10. Overview of the incentives of S2T&B partners ... 67

(7)

List of acronyms

AAWSA Addis Ababa Water and Sewerage Authority

ATWSSSE Adama Town Water Supply and Sewerage Service Enterprise EHNRI Ethiopia Health and Nutrition Research Institute

FDI Foreign direct investment

FDW Fonds Duurzaam Water/Sustainable Water Fund

GoE Government of Ethiopia

GoNL Government of The Netherlands

GTPII/2 Growth and Transformation Plan II/2

JMP Joint Monitoring Programme

MDGs Millennium Development Goals

MNCs Multinational corporations

MoEFCC Ministry of Environment, Forestry and Climate Change MoWIE Ministry of Water, Irrigation and Electricity

NRW Non-revenue water

OWNP OneWaSH National Program

PPP/s Public-private partnership/s

PSI Private sector involvement

RVO Rijksdienst voor Ondernemend Nederland/Netherlands Enterprise Agency

S2T&B Source to Tap and Back

SDGs Sustainable Development Goals

WaSH Water, sanitation and hygiene

WIF WaSH Implementation Framework

(8)

Chapter 1

Introduction

“Water is truly at the core of sustainable development. It is inextricably linked to climate change, agriculture, food security, health, equality, gender and education, and there is already international agreement that water and sanitation are essential to the achievement of many sustainable development goals”

- Michel Jarraud Chair of UN-Water and Secretary General of WMO, in UNESCO (2015, p. vi)

Introduction: the linkages between WaSH, poverty, sustainable

development and partnerships

Around the world, more than 748 million people lack access to an improved drinking water source (almost half of them living in Sub-Saharan Africa) and water demand for industrial manufacturing is expected to increase by 400 per cent between 2000 and 2050 globally (UNESCO, 2015, p. v). Only four years ago, 2.5 billion people did not have access to an improved sanitation facility (WHO & UNICEF, 2014, pp. 8, 12). Today, almost one-fifth of the world's population -about 1.2 billion people- live in areas where water is physically scarce, and one quarter lives in developing countries that face water shortages due to weak governance, human capacities and lack of infrastructure (UNESCO, 2015, p. 19).

(9)

The access to safe water and sanitation facilities is a basic human right1_{and crucial for people} to move out of poverty. Yet, millions of people across the world do not have access to it (see Figure 1). Water, sanitation and hygiene (WaSH) sectors are sectors that demand large investments, extensive infrastructure, technology and capacity for maintenance and operations. Such needs are often troublesome to address, especially for governments across developing countries due to large finance gaps, the high-risk environment that deter (foreign) investment, and lack of skilled labour force. Therefore, many governments seek alternative engagement forms with non-governmental actors to help meet the needs of the WaSH sector. One such engagement is partnering with domestic or foreign private sector. Private companies can bring financial injections, contribute with technological and innovative capabilities and operational efficiencies into the sector, thus help curb the existing challenges the government faces. The public-private engagement often takes shape of multi-stakeholder partnerships or so-called public-private partnerships (PPPs)2_{. These partnerships convene} public, private and civil society actors under different contractual agreements in their joint efforts to tackle pressing development challenges. A range of donors and policy makers support PPPs, and argue that they can contribute to economic growth through job creation, technology transfer, capacity development, innovation, and that they stimulate investment in local economies with financial injections (Hoang Gia & Fugelsnes, 2010; Jamali, 2004; Kauffman & Pérard, 2007; Marin, 2009). Critics argue that PPPs might result in services at higher cost, of poorer quality and limited quantity for marginalised groups (Barlow, Roehrich, & Wright, 2013; Jamali, 2004; Queiroz, 2007).

This thesis seeks to contribute to our understanding about how PPPs in WaSH sectors in developing countries can contribute to sustainable service provision. More specifically, it performs a case study analysis of how one WaSH PPP in Addis Ababa, Ethiopia, is influenced by a number of pre-determined concepts and external factors (see 2.4). But, in order to explain the knowledge gap this thesis aim to fill (see 1.2), formulate the focus and objectives of this thesis accurately (see 1.3), and justify the research questions that guide the study (see 1.4), some background on the topic is required. The following sub-sections briefly explain the linkages between WaSH and poverty, WaSH and sustainable development, and partnerships in WaSH sectors.

Figure 1. Facts about current WaSH challenges in the world

Source: the author, adopted from UN Water GLAAS report 2014 (WHO 2014)

1_{The United Nations Resolution ‘A/RES/64/292 The human right to water and sanitation’ is found in} full here: http://www.un.org/en/ga/search/view_doc.asp?symbol=A/RES/64/292

2_{PPPs are sometimes referred to as Type II agreements, a definition that was developed during the} Johannesburg UN World Summit of Sustainable Development (WSSD) in 2002. Type II agreements aimed to facilitate the inclusion of civil and private actors into the management of sustainable development. Other names for PPPs often used include P3, 3P or, more broadly, multistakeholder-partnerships.

(10)

WaSH and poverty

The link between WaSH and poverty is evident. The daily struggle for water is one of the major burdens of poverty and the reality for many, in particular, women and girls who are often responsible for fetching water to their households

(UNESCO, 2015). In many countries facing water challenges, sources of water in rural areas are commonly found in lack of proximity to where people live and the water is frequently of low quality or unaffordable. Those whose livelihood are dependent upon rain fed agriculture or the availability of water for grazing animals are amongst the most vulnerable to greater water insecurity (Water Governance Facility, 2012).

In urban settings, informal water vendors are sometimes the only source of water, which often exclude the poorest of informal settlements because of unaffordable and

unregulated prices (Water Governance Facility, 2012). Besides, the urban poor often live at the low-pressure ends of distribution networks and cannot afford coping equipment, such as private wells, roof tanks or filters, and are thus disproportionately affected by unsufficient, insecure or low quality service provision (Marin, 2009).

Both in urban and rural environments, not having access to safe water and sanitation facilities can lead to water related diseases and time loss due to cumbersome efforts to fetch water or reach sanitation facilities. This subsequently results in loss of educational and employment possibilities. It also means that people face though choices between paying for water, school fees or medicines which can be critical for families’ health and social dignity (UNESCO, 2015). The access to sanitation facilities might not have such direct influence on life and death as water does. Yet, not having access to sanitation facilities severely implicates the lives for those without. Non-existent or unsafe sanitation facilities are causing diseases, social stigma and environmental pollution. Therefore, the provision of safe water, sanitation facilities and hygiene practices are crucial public services in any society.

WaSH and sustainable development

The development of WaSH sectors and sustainable development are inseparably interrelated. The Chair of UN-WATER Michel Jarraud says “there is already international agreement that water and sanitation are essential to the achievement of many sustainable development goals” (UNESCO, 2015, p. vi). The Director-General of UNESCO Irina Bokova says in agreement that “water is inextricably linked to the development of all societies and cultures” (UNESCO, 2015, p. v). These two front figures of the UN emphasise the important link between WaSH and sustainable development for good reasons. Poverty-oriented WaSH interventions have shown to have direct, immediate and long-term social, economic and environmental results when addressing the issues outlined in the previous sections (UNESCO, 2015). Thus, the lack of access to WaSH services is not only about individual hazards (such as illness or time loss) but can also result in fewer livelihood opportunities, disturbed income generation and lower contribution to economic activity. Water is central to food and energy security, to increasing urbanisation rates and to climate change (UNESCO, 2015), all of which are central factors for sustainable development.

Even though the linkages between WaSH and sustainable development are evident, they are many and often complex. Poverty itself can negatively impact the management of WaSH

“Water is the most

important; water is life.

We cannot choose not

to pay for water”

Resident of Kirkos, Addis Ababa, November 2015

Figure 2. Testimony from the people of Addis Ababa

(11)

services, for instance through unsustainable use of water resources or inadequate technologies. Poverty can also reduce the efficiency of existing investments since households and users or local authorities often find it hard to finance, operate and maintain technical infrastructure or financial billing systems. Estimations show that as much as 30 percent to 50 per cent of water supply projects fail after two to five years (UNESCO, 2015, p. 19) (see 3.4 and 3.5).

WaSH, sustainable development and partnerships

Overcoming WaSH challenges across the world is considered fundamental in order to reach sustainable development. Its importance shows in the Sustainable Development Goals (SDGs)3 and, in particular, Goal No 6 that aims to ensure access to water and sanitation for all by 2030 (UN, 2016). In order to successfully implement the Goals, the SDGs emphasise the importance of multi-stakeholder partnerships.4_{Partnerships as instrument for SDG} implementation are not only advocated by the UN, but increasingly seen as a central component for tackling WaSH challenges and achieving sustainable development by policy makers, donor and beneficiary governments, research communities and other development professionals (see Chapter 3 in full and, in particular, 3.4.).

The importance of WaSH services for poverty reduction and sustainable development is undisputable. Yet, there is no unanimous idea about how to best tackle WaSH challenges through the application of multi-stakeholder partnerships or PPPs. The private sector plays a central role in many bilateral and multilateral donor strategies and engagement with private sector in service provision is emphasised by many low-income countries through their national development plans or sector-specific policies. The following sub-chapter clarifies the problem statement and the gap in knowledge this thesis aims to address.

Problem statement

The problem statement of this thesis involves two main issues with PPPs as an instrument to implement the SDGs. First, the emergence of multi-stakeholder partnerships during the last decades indicates a shift from what has traditionally been understood as responsibilities of the state to an idea that different actors are stronger when collaborating. Vast amounts of research have been carried out on PPPs (see Chapter 3) but despite extensive research, the literature is highly fragmented. This fragmentation is a result of studies applying distinctive problem definitions, diverse methodologies and using data in different ways (see 3.3). Thus, no uniform conceptualisation of the concept of PPP has been commonly accepted but many policy makers, financial institutions and researchers continue to support and encourage the use of PPPs. This causes problems with regards to our basic understanding about the concept, its’ positive and negative impacts and how it should be used, and subsequently, the extent to which we can advocate its usefulness and suitability for addressing development challenges since we, due to the fragmented picture, might not have been able to carry out accurate analyses of previous PPPs. This lack of a clear conceptualisation is the theoretical part of the problem statement.

3_{See http://www.un.org/sustainabledevelopment/(accessed on 13 May 2016).}

(12)

Second, there is genuine concern about the implementation of the SDGs. Its predecessor, the Millennium Development Goals (MDGs), have accomplished substantial progress in development by, amongst other accomplishments, contributing to moving more than 1 billion people out of extreme poverty (United Nations, 2015).5_{Yet, the progress has not been linear} nor universal and all MDGs were not achieved. Thus when the MDGs reached its end in 2015, the persistent challenges were inherited by the SDGs and the Post-2015 agenda. Thus, the discussion about how to implement the Goals has gained rightful toehold. Hajer et al. express that the SDGs “have the potential to become a powerful political vision that can support the urgently needed global transition to a shared and lasting prosperity” (Hajer et al., 2015, p. 1651), but simultaneously problematise the illusion that top-down steering by governments and intergovernmental organizations alone can address global problems. There are also concerns that the SDGs might not be relevant to all countries or interests of all stakeholders, or how quantitative international targets can be reconciled with the qualitative improvements in peacekeeping, state building and the strengthened institutions needed to underpin poverty reduction (Kharas & Steven, 2012). Conclusively, there is little concrete guidance on implementation strategies for the SDGs. Yet, the support to PPPs as a viable instrument for implementing the Goals continues. The ground we are currently breaking as we embark on the implementation of the Post 2015 agenda is marked by uncertainty and forms the empirical part of the problem statement.

The theoretical fragmentation of PPPs but its’ persistent support coupled with the centrality of PPPs in the SDG implementation agenda has inspired the direction of this thesis. This thesis aims to analyse whether these factors have implications for PPPs in practice by performing an instrumental analysis of the concept (see 2.4). Based on such analysis, it aims to contribute to the debate about whether a popular instrument such as the PPP is indeed a viable implementation instrument of the SDGs. If they are, which are the crucial factors in partnership design that can enable its contribution to sustainable development. Thus, the thesis aims to create a bridge between the theoretical fragmentation of the concept and the empirical experience of PPPs. With that, the thesis contributes with valuable knowledge directed to policy makers, donors and PPP practitioners.

Focus of study

This thesis was produced at an interesting and critical time in the international development agenda. The successes and failures of the Millennium Development Goals (MDGs) have been evaluated and the SDGs are currently in its initial implementation phase. Discussions about the potential and possible pitfalls of the SDGs are characterising much of the international debate about development, and the viability of its implementation strategies (or lack thereof) are examined.

The scope and limits of an MA thesis demands specific focus and methodological and thematic choices in order to increase the feasibility and quality of the study. This thesis applies a specific focus in two ways. First, it acknowledges the framework of the SDGs as a scope within the broader development agenda (although recognising also the broadness of the SDGs itself). Second, it further narrows down its focus to analysing PPPs in the WaSH sector exemplified through a case study (see 2.6). Thus, its primary focus targets two of the SDGs,

5

See the United Nations (2015) ’The Millennium Development Goals Report 2015’ for a full overview of the accomplishments.

(13)

Goal 6 “Ensure availability and sustainable management of water and sanitation for all” and Goal 17 “Strengthen the means of implementation and revitalize the global partnership for sustainable development”. More specifically, it analyses how PPPs can best be designed in order to contribute in meeting the targets of Goal No 6 though instruments encouraged in Goal No 17.

Research questions

The overall research question that guides the study has been developed based on the theoretical understanding and empirical experiences of PPPs and its’ continuous support by many actors in the Post-2015 agenda (see 1.2 and 2.5). Hence, the main research question is:

RQ

How can PPPs be designed in order to contribute to sustainable urban water and

sanitation services?

In order to answer this question, the study applies a revised version of Young’s (2005) framework for institutional analysis (see 2.4). The framework helps formulate a number of sub-questions (outlined below) and is combined with a case study research design that allows an investigation of one PPP in Addis Ababa, Ethiopia (see 2.6). The sub-questions address different parts of the conceptual framework (see 2.4 and Table 2) and have been formulated as follows:

SQ 1

Which are the institutions that influence the design of the PPP, how does such influence look like, and how compatible are the institutions with each other?

SQ 2

What incentivizes the different partners to participate in the PPP and how do they understand the the concept of PPPs?

SQ 3

What is the user experience of the WaSH services in Addis Ababa and how does the PPP address potential issues the users face?6

SQ 4

Which are the driving forces of the Ethiopian WaSH sector, i.e. the underlying reasons for the persistent WaSH challenges?

SQ 5

How do the (anticipated) impacts of the PPP address issues stemming from the driving forces?

SQ 6

Do we need to redesign PPPs in order to improve its’ potential contribution to sustainable urban WaSH services? If yes, how?

6

This sub-question has at a later stage of the research been excluded from the study. See 2.4 for an explanation of the exclusion.

(14)

Organisation of the thesis

The thesis consists of six chapters and a number of appendices. Each chapter represents a different step of the research process and discusses the findings in different sub-chapters. This first chapter, Chapter 1, has introduced the topic of the thesis (see 1.1), explained the problem statement and the contribution of the research (see 1.2), briefly discussed its focus (see 1.3), and presented the research questions (see 1.4). Chapter 2 concerns the research methodology applied. Chapter 3 presents the literature review that forms the theoretical foundation of the study. Chapter 4 and 5 discuss the research findings by following the order of the conceptual framework: Chapter 4 starts with an overview of WaSH and the business environment in Addis Ababa, and examines the role of institutions, incentives and driving forces, thus addressing sub-question 1-4. Chapter 5 builds on the findings of the previous chapter and examines the impacts and redesign of PPPs, thus addressing sub-questions 5 and 6. The final chapter, Chapter 6, provides some concluding remarks and outlines recommendations for partnership design. It also reflects on the overall study and future research. The content of each chapter is presented more detailed in the respective introductory sub-chapters.

(15)

Chapter 2

Research methodology

Introduction

This chapter explains the methodological choices and research design of the study. The chapter follows the order of how each step of research was carried out, starting by introducing the ontological and epistemological point of departure (see 2.2) and the literature review (see 2.3). Based on the literature review, the conceptual framework was developed (see 2.4) and the context and scope of the study (see 2.5) determined. At this stage of research, a case study was considered the most appropriate choice of research design (see 2.6). As part of the research design, the exploratory ambitions of the study are presented (see 2.6.1) followed by the data collection techniques consisting of access and sampling strategy (see 2.6.2), documents, (see 2.6.3) and interviews (see 2.6.4). The data collection phase was later followed by the data analysis (see 2.7). The two last sections of this chapter discuss the ethical considerations (see 2.8) and quality criteria of research and methodological limitations (see 2.9).

Epistemological and ontological position

This thesis departs from a critical realist epistemological and ontological understanding that there is a difference between the existing reality and the conceptions and understanding of the same. This implies that some entities (such as reality, or ‘the world’) exist independently

(16)

of the perception or understanding of them. Thus, the existing reality can be interpreted in multiple ways depending on the available knowledge (Sayer, 2006).

The thesis acknowledges the three overlapping domains of reality of critical realism (see Figure 3). The basic assumption of reality is that generative mechanisms in the real domain, in combination with contextual factors and events in the actual domain, influence observable effects and outcomes in the empirical domain (Bryman, 2012; Sayer, 2006). Bryman (2012) explains the interrelations between the three domains by saying: “the identification of the context that interacts with the generative mechanism to produce an observed regularity in the social world” (Bryman, 2012: 29). With regards to this thesis, critical realism corresponds well to the choice of a case study research design since case studies aim to draw inferences between different perspectives or ideas (Bryman, 2012).

Source: the author, based on Bryman, 2012; Mingers, 2004; Sayer, 2006.

Literature review

In any study, the literature review serves the purpose to understand what is already known about the topic at hand. It helps to better comprehend what concepts and theories have been applied, what methods have been used, what controversies exist about the topic, what clashes of evidence (if any) exist, and who the key contributors to research on the topic are (Bryman, 2012). All of these points support the researcher to identify a gap in knowledge that will inform the direction of the study and to demonstrate the credibility and contribution of the research. An adequate literature review thus guides the researcher to the most appropriate ways -theoretically, methodologically, ethically and so on- to fill that knowledge gap. It also helps the researcher to find support to the research questions, the findings, the conclusions and potential recommendations of the research.

The literature used in this study was determined by a number of key words and concepts typed into open search engines, including Google and Google Scholar, and the university Figure 3. Three overlapping domains of reality of Critical Realism

(17)

library search functions of the University of Amsterdam (see Table 1). The literature reviewed include studies on: the rise of PPPs in WaSH service provision (see 3.2) including a discussion about the value of WaSH (see 3.2.1), the challenge of conceptualising PPPs (see 3.3), and the impacts of PPPs in a developing context (see 3.4) including factors that contribute to those impacts (see 3.5).

A literature review does two things. First, it forms the theoretical and scientific foundation that enables larger inferences about the role of PPPs as an instrument for implementing sustainable WaSH services in an urban developing context. Thus, it enables a discussion about the role of PPPs in the global agenda for sustainable development. Second, the review of previous studies appropriately informs the case study selection (see 2.6), which in turn informs the search for relevant respondents (see 2.6.2).

Table 1. Overview of key words and concepts used for literature search

KEY WORDS AND CONCEPTS USED FOR LITERATURE REVIEW

Public-private partnerships PPPs

Partnerships Water and sanitation WaSH sector

Private sector development Private sector involvement

Private sector engagement in WaSH Private sector public service provision Driving forces of Ethiopia WaSH Drivers of WaSH

SDGs

Sustainable WaSH services Addis Ababa Ethiopia Sub-Saharan Africa

Search engines: Google, Google Scholar, CataloguePlus (University of Amsterdam) All key words/concepts have been combined in multiple ways

Conceptual framework

After reviewing the literature, the conceptual framework was developed. Determining and using concepts in social science research serve several purposes. They are essentially labels that we give to aspects of the social world that seem to share common features and strike us as significant. Concepts also help us organise research processes and research findings and facilitate the message we want to send to our intended audience (Bryman, 2012).

The conceptual framework guiding this thesis was originally created to study the influence of institutions and their effects on global environmental change (see Young et al., 2005). In the IDGEC7_{Science Plan, Young argues that institutions can play a large role in both causing and} addressing large-scale development changes and challenges. In order to further examine this

7_{The Institutional Dimensions of Global Environmental Change is a long-term international research} project developed under the auspices of the International Human Dimensions Programme on Global Environmental Change (IHDP). For more see Young et al., 2005.

(18)

role of institutions, the Science Plan applies a sequential research foci consisting of three steps combined with the conceptual framework for analysis.

This thesis applies Young’s conceptual framework, but in a revised version (see Figure 4). The framework has been revised in order to better correspond to the overall aims of this thesis (see 1.3.). Methodologically, the revision helps carrying out sequential research by making the different phases of data collection and data analysis follow the order of the conceptual framework (see 2.6 and 2.7). This helps keeping the research process on track and facilitates informed methodological choices as the research process goes on.

Figure 4 presents the revised framework and

Fourth, concept C is marked in grey in Figure 3 and Table 2, which indicates that it has been excluded from this study. The exclusion is based on two factors. First, at the time of the data collection the focus of the study was sustainable water services, excluding the sanitation part of the system chain. Therefore, the questions in the questionnaire only targeted the users’ experiences with water services and the results are naturally not depicting the whole WaSH situation. However, the concept is very important as it is the only concept that incorporates the view of the users. The user perspective in studies like this is complex and deserves full focus departing from their role at the core of the study. Such focus should be applied in future research (see 6.2). Second, the limited scope of a MA thesis demands choices and limitations (see 2.5). The interested reader can consult Appendix 5: Original set-up of SQ 3that contains the work that was carried out for the analysis of concept C, although its’ findings are not incorporated in the overall thesis.

(19)

Table 2 further explains how each concept has been revised, presents the indicators that are used to analyse each concept (i.e. the operationalisation of the study), and justifies the exclusion of Concept C.

Figure 4. Conceptual framework for instrumental analysis: concepts and relations between concepts (revised from Young et al., 2005)

The following remarks outline the main reasons for the changes of the framework.8

First, this thesis aims to analyse how PPPs can be designed as an instrument for sustainable WaSH services. Such objective puts the PPP –the instrument- at the center of the study. Similar to Young’s framework, the thesis departs from the assumption that instruments are influenced by (amongst other concepts) institutions. However, the shift from Young’s core focus on an institutional analysis to the focus on an instrumental analysis thus fits the overall research objective better.

Second, by putting the instrument/PPP at the center of the study, the framework enables an investigation of what incentives to partake in a PPP the partners seem to have. Partner compatibility is a crucial factor for successful partnership outcomes (see Chapter 3), and therefore important to investigate in order to answer the overall research question (of how to design PPPs in order to contribute to sustainable WaSH services).

8

Appendix 2 gives a comprehensive overview of how the conceptual framework has been revised and for what reasons.

(20)

Third, this thesis aims to contribute with evidence-based policy-oriented research. In such research endeavour, it is central to focus on concepts that are important to policy makers, relevant stakeholders or implementers. If not, the research findings and recommendations have lower chances of being considered. Therefore, the framework has been revised in order to enable a deeper analysis of two steps starting with: i) which are the underlying driving forces that cause the existing challenges of the WaSH sector in Addis Ababa (see 4.5), and ii) how the PPP in its current form addresses those challenges or not (see Chapter 5). Given the answers to these questions, we can determine what alterations are needed in the design of PPPs in order to enhance their contribution to sustainable development (see Chapter 5 and 6). Such analysis will, hopefully, speak louder to policy makers and implementers as they can only bring about changes when instruments (the PPP) appropriately target the underlying driving forces of persistent development challenges.

Fourth, concept C is marked in grey in Figure 3 and Table 2, which indicates that it has been excluded from this study. The exclusion is based on two factors. First, at the time of the data collection the focus of the study was sustainable water services, excluding the sanitation part of the system chain. Therefore, the questions in the questionnaire only targeted the users’ experiences with water services and the results are naturally not depicting the whole WaSH situation. However, the concept is very important as it is the only concept that incorporates the view of the users. The user perspective in studies like this is complex and deserves full focus departing from their role at the core of the study. Such focus should be applied in future research (see 6.2). Second, the limited scope of a MA thesis demands choices and limitations (see 2.5). The interested reader can consult Appendix 5: Original set-up of SQ 3that contains the work that was carried out for the analysis of concept C, although its’ findings are not incorporated in the overall thesis.

(21)

Table 2. Operationalisation of concepts and their respective indicators

DIMENSION OF CONCEPT INDICATORS

A

The SDGs / Goal #17 Means of influence and expectations of private sector involvement (PSI) and PPPs for sustainable development

Compatibility with the other institutions

The Netherlands Government /

‘Aid and Trade’ policy Same as above

Ethiopian Government / ‘GTP2’ policy

Same as above

B

PPP and partner incentives Understanding of partnership

To be determined by open coding To be determined by open coding

C

Users / beneficiaries Satisfaction with current water service and willingness to pay more for improved

service

D

Physical driver Financial driver

Geographical location and existing infrastructure Financial gaps in the sector

Technological driver Technology transfer and capacity development

Demographical driver Health problems and rapid urbanization

E

Infrastructural improvement Financial injections Technology transfer/capacity development Health-oriented interventions Positive (+) / negative (-) +/- +/- +/-

F

Recommendations redesign of the instrument (PPP) to enhance its contribution to sustainable WaSH services

Recommendations based on research findings

Research context and scope

This thesis is part of the broad and complex field of sustainable development. More specifically, it is part of the discussion about implementation instruments that will help achieving the SDGs. Even more specifically, it analyses one type of implementation instruments, namely the widely supported format of PPP (see 1.2). Any type of research demands a certain level of specificity; it is necessary to zoom in on a specific context to give weight and relevance to the research findings and recommendations, and to enhance the quality of the research. It is impossible to cover all aspects or perspectives of a given topic (Bryman, 2012). In addition to the research questions (that naturally determines the scope of the thesis, see 1.4), some choices concerning the research context are explained below.

(22)

The scope of this thesis is to perform an analysis of one specific PPP that operates in the WaSH sector in Addis Ababa, Ethiopia (see 4.5.1, Figure 17 and Figure 18 for an overview of the case study location, and 4.4.1 for an explanation of the PPP). The choice of, for instance, conducting a comparative study analysing different PPPs does not fit the research questions this thesis has set out to answer. Instead, the specificity of one project enables a better understanding of the particularities of that given PPP and the potential influence of context specific factors. Yet, analysing a specific project in full is an ambitious task with a wide range of noteworthy factors or perspectives. Therefore, this thesis limits the analysis of the PPP to the concepts found in the revised conceptual framework (see 2.4).

Concerning the choice of which PPP to analyse, the selection was driven by the following considerations with regards to the PPP itself:

a) the selected PPP aims to tackle existing water challenges in Addis Ababa in order to improve water quality and services to the population, i.e. to contribute to achieve clean water and sanitation for all (SDG 6), and is thus a good example of the broader field this research is part of;

b) the selected PPP is considered representative for the larger FDW subsidy programme by its composition of public and private partners, target area and project objectives, and for a larger amount of development PPPs initiated by major donors aiming at tackling challenges in public service provision (see 3.2);

c) previous research on PPPs in WaSH sectors is vast and can therefore support the case study (see Chapter 3 in full).

The case study location was selected based on: i) Addis Ababa provides an interesting set-up for analysing urban water and sanitation projects aiming at improving services given the current challenges of Ethiopia in general, and Addis Ababa in particular, in reaching sustainable WaSH services (see 4.5), and ii) Ethiopia has limited experience with foreign private sector involvement in general, and in the format of PPPs in particular (see 4.2). These factors combined makes for an interesting and relevant case study.

Research design: case study

A case study research design aims to generate an intensive examination of a single case (Bryman, 2012). It provides a framework for data collection and data analysis and allows an analysis of the complexity and particular nature of the case in question. As part of any case study research design, it is useful to highlight the type of case the study is analysing. By distinguishing different types of cases we can easier understand both strengths and limitations of the research design (Bryman, 2012).

Following the case distinction from Yin (2014), which is later revised (and renamed) by Bryman (2012), the case chosen in this thesis is best described as an exemplifying case. Such case can be used to exemplify a broader category of which it is a member. In this study, the case is a PPP funded by public donor money aiming to tackle water challenges. In such category there are many similar cases to choose from. However, it is important to note that the contexts in which the similar cases (publicly funded PPPs in water sectors) operate significantly differ from each other. Therefore, the exemplifying strength of one case with regards to making wider inferences about its category might be limited (Bryman, 2012; Yin,

(23)

2014). A strength can be, however, to instead draw inferences to publicly funded PPPs in other sectors at the case study location (but that is not an aim of this study).

Research set-up

The research was conducted following five phases (see Figure 5) starting with a preparatory phase consisting of a literature review, case selection and initial search for respondents. This phase was then followed by three subsequent phases of data collection that took place in The Netherlands and Ethiopia. The final data analysis was the last phase. However, the initial data analysis started during Phase I, and continued throughout the subsequent phases II and III. Conducting initial data analysis during data collection helped determine appropriate and unforeseen research paths that changed subsequent data collection strategies. For instance, the document analysis helped formulate interview questions (see 2.6.3) (Bryman, 2012). Ethical considerations were taken into account throughout all different phases (see 2.8).

Figure 5. Phases of research

2.6.1 Exploratory research

This thesis has exploratory ambitions in that sense that it aims to potentially explain how PPPs can best be designed to contribute to sustainable WaSH services given certain influencing factors. The choice of conducting exploratory research is connected to the epistemological and ontological point of departure. As critical realism acknowledges and emphasises the causal mechanisms generating actual and observable events, a critical realist approach is suitable for exploratory research rather than predictive research (Bryman, 2012). The research design does not involve any predictive ambitions.

(24)

Following critical realism, the appropriate data collection strategies were determined by their qualitative character since the study aims to understand context specific circumstances (that can take shape in observable or unobservable events) (Bryman, 2012; Sayer, 2006).

2.6.2 Access and sampling strategy

Access to key stakeholders or necessary informants can be challenging for any case study, and the lack thereof might have detrimental implications for the research. Therefore, establishing contact with relevant organisations or individuals at an early stage of the research process is crucial (Bryman, 2012). Gaining access to key respondents in the WaSH sector in Addis Ababa was challenging. Before the data collection started, the project coordinator of the PPP was contacted and expressed a positive approach towards the study. Thanks to this positive approach, necessary contacts were established at an important stage of research. During the data collection, the same individual introduced the remaining PPP partners and thereby granted access to expand the number of respondents interviewed (see 2.6.4). Since the case study is a specific PPP with a number of given project partners, strategic and purposive sampling of interview respondents was a necessary strategy. The respondents that were not partners of the PPP were similarly purposely approached based one their relevance for the WaSH sector (see 2.6.4). The non-probability form of purposive sampling places the research questions at the heart of sampling considerations rather than sampling research participants on a random basis that in turn increases the generalizability of the results (Bryman, 2012).

2.6.3 Documents

The thesis has collected and analysed a range of documents. The documents are divided into two categories: i) primary documents that were used for analysing the influence of institutions (concept A), thus answering sub-question 1, and ii) secondary documents that were used to analyse remaining concepts and sub-questions and serve as complement to interviews. Concept A was analysed based on documents from two levels of institutions (see 2.4): i) global level through the SDGs (see 4.3.1), and ii) national level through the development policy of The Netherlands (see 4.3.2) and the national development plan of Ethiopia (see 4.3.3). There was an initial ambition to incorporate PPP project documents but due to their confidentiality, this was difficult to realise. The selection of the specific documents was based on their centrality for the respective institution with regards to their influence on policy and implementation.

(25)

Table 3. Overview of documents

ORGANISATION NAME PART OF DOCUMENT ACCESS

GLOBAL LEVEL

United Nations A/RES/70/1 - ‘Transforming Our

World: the 2030 Agenda for Sustainable Development’ /SDGs

Goal #17 ‘Partnerships for the Goals’, para. 17.16 and 17.17

Online (see UNGA, 2015)

NATIONAL LEVEL

The Netherlands, Ministry of Foreign Affairs

‘A World to Gain – A New Agenda for Aid, Trade and Investment’, April 2013

Document in whole Online (see Ministry of

Foreign Affairs of the Netherlands, 2013) Ethiopia, National

Planning Commission

‘Growth and Transformation Plan II 2015/16-2019-20’, May 2016

Document in whole Via respondent G1

AAWSA Water Balance Model Document in whole Via respondent G3

S2T&B Project documents Available project documents Confidential

Documents used for the institutional analysis of concept A

Note that the initial document analysis was carried out before the interviews took place and therefore helped determine interview topics and questions. How the document analysis was performed is further explained in sub-chapter 2.7.

2.6.4 Interviews

Respondents

The bulk of the data was collected through semi-structured interviews with respondents based in The Netherlands and Ethiopia. Most interviews were held during the 2,5 months’ period of data collection in Addis Ababa. The majority of the interviews took place in person, with exceptions of a few that were conducted via Skype. The respondents were eligible based on either being representatives of the partners in the PPP or being knowledgeable in the WaSH sector of Addis Ababa or in private sector development. Five of nine PPP partners were interviewed (see 4.4.1). The total amount of 31 respondents were from the following categories:

(26)

26 UNIVERSITEIT VAN AMSTERDAM Table 4. Overview of types of respondents and respondent categories

DONORS (D)

Bilateral (4)

Embassy of The Netherlands Ethiopia (2) Netherlands Enterprise Agency (RVO) (1) Embassy of Sweden (1)

Multilateral and financial institutions World Bank (3)

UNDP (2) UNICEF (1) UNECA (2)

ETHIOPIAN GOVERNMENT (G)

Ministry of Water, Irrigation and Electricity (MoWIE) (2)

Ministry of Environment, Forestry and Climate Change (MoEFCC) (1) Water Resources Development Fund (WRDF) (1)

Addis Ababa Water and Sewerage Authority (AAWSA) (2)

Adama Town Water Supply and Sewerage Service Enterprise (ATWSSSE) (1)

PRIVATE SECTOR (P)

Vitens Evides International (VEI) (1) MetaMeta Holding B.V. (1)

Private consultants in WaSH/PSD in Ethiopia (2)

NON-PRIVATE SECTOR (N)

WaterAid Ethiopia (3)

KNOWLEDGE INSTITUTIONS (R)

Ethiopian Health and Nutrition Research Institute (EHNRI) (4) PPPLab The Netherlands (1)

Interview format

The main part of the interviews followed a semi-structured format. They were performed with a predetermined interview guide with determined topics and/or questions. Most respondents were informed about the content and aim of the interview beforehand, either through communication via email or by telephone. A number of interviews were conducted in a more informal manner (such as information sharing in an office corridor, or informal talks at restaurants/bars) and were therefore performed slightly different. All interviews except for one were conducted in English. The one was instead conducted in Amharic and the local supervisor acted as translator.

(27)

Data analysis

The data analysis serves multiple purposes. It manages raw data, identifies possible flaws in the data collection, and performs data reduction by structuring and organising the data. This means that the steps of data analysis are concerned with reducing the large amount of data so the researcher can make sense, and use, of it. Depending on the type of data that has been collected through the different methods, the steps of data analysis take different shapes. Therefore, it is important to identify what type of data you are dealing with before the analysis begins (Bryman, 2012).

The the data in this thesis is of qualitative character collected from documents, and interviews. In practice, such data consists of a large amount of unstructured textual material and is therefore not too straightforward to analyse (Bryman, 2012). The computer software ATLAS.ti was used due to its suitable features of identifying patterns, concepts, and linkages within datasets. The computer software Excel was used for descriptive statistics.

In order to manage, analyse and interpret the data, the following steps of data analysis were carried out. Table 5 below gives an overview of the coding scheme (the different coding strategies are further explained in the following sub-sections).

Table 5. Coding scheme

RELATED SUB-QUESTION

DATA CONCEPT DIMENSION CODE*

(examples, not exhaustive) SQ 1

Which are the institutions that influence the design of the PPP, how does such influence look like, and how compatible are the

institutions with each other?

Documents

Interviews #influencing institution _means of influence _expectations #inflinst_NL _mean_finance to Dutch private sector #infl inst_ETH_exp_ lack of foreign currency

SQ 2

What incentivizes the different partners to participate in the PPP and how do they understand the concept of PPPs?

Interviews #PPP

#incentives _service delivery _responsibility /accountability _market expansion _reputation/CSR #incentive_continuous service delivery #incentive_accountability #incentive_responsible actor #incentive_priv sec seek long-term profit through market expansion #incentive_reputation of actor #incentive_part of CSR portfolio

SQ 3

What is the user experience of the WaSH services in Addis Ababa and how does the PPP address potential issues the users face

Questionnaires #users _service satisfaction

_willingness to pay No codes, instead descriptive statistical analysis

SQ 4

Which are the driving forces of the Ethiopian WaSH sector, i.e. the underlying

Interviews Documents

#driving forces _physical _financial _technological _demographical

#driver_PH_supply does not meet demand

#driver_FI_finance gaps

(28)

reasons for the persistent

WaSH challenges? #driver_DE_rapid urbanisation

SQ 5

How do the (anticipated) impacts of the PPP address issues stemming from the driving forces?

Interviews

Documents #impacts _physical +/- _financial +/- _technological +/- _demographical +/-

#impacts_PH_reduction of erosion (50%) in buffer zone

#impacts_FI_Self financing water education centre

#impacts_TE/CA_Water safety plans introduced

#impacts_DE_Reduced occurrence of water borne diseases

* The exemplified codes derive from the selective coding round. Therefore, each code follows the format: CONCEPT + DIMENSION + CODE. In the first open coding round the codes were only CODE

Step 1: Document analysis

The first step of data analysis commenced after the initial phase of the data collection (see Figure 5 in 2.6). An analysis of the selected documents (see 2.6.3) was performed early because it helped identifying additional respondents and constructively informed the design of the subsequent interview guide (see 2.6.4). The analysis of the document was later repeated after the interviews were analysed. This re-analysis was done in order to i) reach saturation with regards to how the information retrieved from the documents contributed to the overall research, and i) contribute to triangulation of the data after combining it with the information retrieved from the interviews (Bryman, 2012). Such iterative process was useful as it allowed for changes in the following data collection phases (Bryman, 2012).

The data retrieved from the documents was of qualitative character. Therefore, the analysis aimed to identify indicators of how the selected institutions (visible through the documents) influences private sector involvement or PPPs. As the documents are quite generic by nature (they are the most overarching plans or policies of their respective domicile), they do not go much in-depth into specifics. Therefore, the documents were analysed using selective coding (Bryman, 2012). The pre-determined core categories were identified to “means of influence” and “expectations” (see 4.3). The identification of those categories helped understand how the influence of the institutions looks like, thus answering sub-question 1 (see 1.4).

Step 2: Interviews

The notes from the interviews were transcribed immediately, or within the day, after it was performed. In this way, the data generated is as thick as possible and the risk of loosing important information or nuances is lower (Bryman, 2012).

The data retrieved from the interviews was of qualitative character and represented the bulk of the collected data. Therefore, in order not to oversee important information, the first round of analysis applied open coding. Open coding is “the process of breaking down, examining, comparing, conceptualizing and categorizing data” (Bryman, 2012, p. 569). This coding strategy allowed for a careful analysis that would allow for new topics, themes or insights to come out. It also allowed for the initial categorisation of the data. This coding resulted in 576 unique codes and 5 code groups (types of respondents, see Table 4). After the first categorisation, a second coding round applied selective coding. This helped strengthen the

(29)

categories (Bryman, 2012). The categories represented the empirical versions of the different concepts in the conceptual framework. The second coding round resulted in 577 unique codes, 5 code groups (types of respondents), 4 concept groups (that gathered information relevant to concept A, B, D, E in the conceptual framework), and several sub-dimensions (see Table 5).

Ethical considerations

Ethical issues are important to consider throughout the whole research process; in case selection, respondent sampling, data collection and data analysis, and after the research project finishes. Bryman identifies the main ethical principles into whether there is: i) harm to participants, ii) informed consent, iii) invasion of privacy, and iv) deception involved (Bryman, 2012, p. 135). These principles are of particular importance with regards to participatory or observatory social research (Bryman, 2012). Since this is not the methodological path of this thesis, the main ethical issues of concern are identified somewhat differently and explained separately below.

Informed consent of participation

Each respondent has been properly introduced to the aims and use of the study and their role in participating. The initial plan was to use written consent forms but after the first couple of interviews, this decision was re-evaluated due to negative experiences of respondents when faced with such paperwork. Instead, verbal consent was used. For similar reasons, handwritten notes were taken during interviews instead of using recording devices. Upon request, some respondents were provided official documentation and approval of the research project from the UvA.

Ensured privacy and anonymity of participation

All respondents have been verbally ensured anonymity and confidentiality of participation to my best ability. Yet, many were not concerned about anonymity but instead pleased to officially participate. The combination of overt methodologies (which means that all respondents have been fully aware of the role of the researcher and their participation in the study) and informed consent means that the risk of invading respondents’ privacy is very low. Still, anonymising data measures have been used. For instance, the transcripts and anonymised by assigning the respondents id codes. These id codes are then used in the thesis when referring respondents’ statements. The id codes only show the capacity of the respondent (e.g. D as in donor, P as in private sector etc.). The names and capacity of each respondent are kept separately.

Deception

Deception occurs when researchers represent their work as something else than it is (Bryman, 2012). It has been central to this thesis to ensure that respondents did not feel decieved in their participation or by the researcher or questions. Therefore, the aim of interviewing respondents of the case study –to better comprehend the specific PPP and how it could contribute to sustainable services- has been clearly communicated to each respondent a priori to their

Designing public-private partnership for sustainable urban water and sanitation services : a case study of Addis Ababa, Ethiopia

Abstract

Acknowledgements

Table of contents

List of figures and tables