Frame analysis of the water problem in Maputo, Mozambique

(1)

(2)

Frame Analysis of the Water

Problem in Maputo, Mozambique

GRADUATE SCHOOL OF SOCIAL SCIENCE MSC THESIS

INTERNATIONAL DEVELOPMENT STUDIES AUGUST 2018

(3)

ABSTRACT

In the year 2000, a flood occurred in Mozambique with catastrophic results. Ever since, the country has faced irregular rain patterns. Since 2015, the government has imposed seasonal restrictions on drinking water supply in Maputo. Several initiatives aiming at solving the problem are currently being implemented or still in the discussion or planning phase. Unlike the research conducted to date regarding this issue, mostly inclined to defining specific causes and solutions to the problem, this study tackles the problem from an interpretative policy analysis perspective, mapping the different prevailing discourses on the water crisis and understanding the interplay between them. During a two-month fieldwork period in Maputo, I employed qualitative data collection methods that consisted of in-depth interviews with stakeholders (donor agencies, NGOs, researchers and government bodies). Information derived from the analysis of project documents, reports and articles provided the necessary context and background. This gave me the ability to scrutinize the different ways in which actors involved in the urban water sector frame the problem, the way it is contested, and how different framings include different solutions.

The study distinguished five different framings, of which three blame the problem on human causes: transboundary water management, urban planning and corruption. One framing, which is that of poor water management, acknowledges the natural processes underway while emphasizing their exacerbation by human factors. The last and most dominant framing is that of climate change, which implies the depoliticisation of the issue as it fully places the problem under natural causes. The proposed solution under this framing in order to overcome the crisis is that of the finalization of a dam, the “Corumana”.

Based on these findings, the study concludes that the climate change framing tends to shift the attention away from the existing managerial issues, mainly those related to the mismanagement of water resources by different stakeholders.

Keywords: frame analysis, water scarcity, Maputo, discourse analysis

(4)

ACKNOWLEDGMENTS

This study took place under the framework of the research project “Sustainable freshwater supply in urbanizing Maputo, Mozambique”. This project is funded by NWO WOTRO (Netherlands Organisation for Scientific Research - Science for Global Development) as part of the Urbanising Deltas of the World research programme (UDW), and is co-implemented by IHE Delft. Scarcity of freshwater threatens the survival of delta cities worldwide. According to an IHE project report, Maputo, which lies in the Incomati delta in Mozambique, is a typical example of an area with freshwater shortage, leading to competition for water and public health hazards. The majority of the population lives in informal settlements not connected to centralised, piped water supply. This project explores whether and how wastewater can be reused on the basis that it can significantly contribute to solving scarcity problems. However, water management organisations currently lack knowledge, tools and capacity to integrate reuse in the overall design of water systems (Rietveld, L., 2013). The project therefore aims to increase fundamental knowledge on water reuse and contribute to developing new technologies and approaches to integrate reuse in the overall water systems. It also aims to produce reliable technologies, tools, models and approaches for local stakeholders who need water reuse strategies and solutions to develop this market. These stakeholders include governmental organisations, non-governmental organisations and private operators; conform to the approaches proposed in the Mozambican Water Policy. In order to conduct this research, the program employs partnerships with Eduardo Mondlane University (UEM) in Maputo where its four PhD students are based (Rietveld, L., 2013).

(5)

PERSONAL

Firstly, I would like to thank my supervisor, Anna, for her continuous support. Anna, thank you for providing me with enough liberty to kick off the writing process, coupled with the necessary guidance and criticism to help me convert my “ideas on a page” into a structured text. Thank you for your encouraging words, I am truly happy to have had a balanced supervision. Secondly, I would like to thank Adriano, my local supervisor, for giving me a comprehensive introduction to the Mozambican water sector and for his availability to discuss and give me advice on my progress throughout the fieldwork. Joaquina, my translator who turned into a research assistant, thank you for your indispensable daily help in conducting interviews as well as logistical support.

Chris, you made me feel right at home the minute you picked me up from the airport. Thank you for your generous hospitality, your Mozambique tips, your delicious cooking and for being a fantastic housemate. I cannot wait to accidentally follow you to yet another country. My Amsterdam family - thank you for making this year a truly unforgettable experience that I deeply cherish, and for sharing the toughest and most enjoyable times together. I wish you happiness and fulfilment in the paths you take from here.

Finally, I would like to thank my parents. Thanks to my precious mother who insisted on accompanying me to Amsterdam, strolling me around the city in a wheelchair. Thank you for your support that manifested itself in different levels and for believing that I should not miss this opportunity, and more importantly, for always believing in me.

(6)

ACRONYMS & ABBREVIATIONS

ADEM Águas da Região de Maputo

AIAS Administração de Infraestruturas de Abastecimento de Água e Saneamento ARA Sul Administração Regional de Águas do Sul

CRA Conselho de Regulação do Abastecimento de Agua

DFATD, Department of Foreign Affairs, Trade and Development (Canada) DFID, The Department for International Development

DNA National Directorate of Water

DNAAS Director Nacional Abastecimento de Agua e Saneamento DNGRH Direcção Nacional de Gestão de Recursos Hídricos

FIPAG Fundo de Investimento e Patrimonio do Abastecimento de Agua GoM Government of Mozambique

IPCC Intergovernmental Panel on Climate Change JICA, Japan International Cooperation Agency PoP, Pequenos Operadores Privados

UNICEF, UN International Children’s Emergency Fund USAID, United States Agency for International Development WSUP, Water & Sanitation for the Urban Poor

(7)

1. Chapter One: Introduction

1.1 An Unfortunate Series of Water Crises

Mozambique is located on the south-eastern coast of Africa and shares 9 of its 13 river basins and aquifers with neighbouring countries. Located downstream from the major rivers in the region, it is at a disadvantage when it comes to natural disasters of different forms as it experiences them for an extended period and in greater intensities. These events have hindered the country’s economic growth on several occasions. The country has experienced three significant floods which devastated the nation in the past two decades: in 2000, 2007, and most recently in 2013 (GFDRR, 2014). Although the flooding in 2013 reached the same level of watershed, it had a lower effect on the population itself compared to the 2000 catastrophic floods, which affected a total of 4.5 million individuals, the poorest being the most impacted. (GFDRR, 2014)

There is a significant presence of international actors in Mozambique; in addition to the agencies that started operating after the end of the civil war in 1992 under the “post-war reconstruction” umbrella, the aforementioned climate events have attracted additional attention and funding. Such factors have raised the country’s national budget dependence on external assistance. According to WASHwatch, institutional donors have spent USD 70 million in the Mozambican water sector alone between the years 2007 and 2016. (WASHwatch, 2018)

In 2015, rainfall was significantly below average in the southern more populous region of Mozambique, and consequently the government imposed drinking water restrictions as a result of the ensuing drought. For the past three years, restrictions on water consumption (for domestic, industrial and agricultural use) have been imposed by the Mozambican state in Maputo and Matola due to the lack of water in the reservoir supplying these cities. In order to alleviate the current situation, various local and international actors embedded in the Mozambican water sector are discussing or already mobilizing support through different funding mechanisms.

(10)

2

1.2 Research Questions

Through this research, I look into the different ways in which actors involved in Maputo’s urban water system understand the problem, and the way it leads to different solutions. To do this, I use a frame analysis. First, I start by getting acquainted with the potential existing framings through various government and NGO documents. After that, in order to identify the dominant framings within the water sector, I consult the different stakeholders involved in the Maputo water system to find how the problem is perceived from their points of view, which groups of individuals it affects the most, which strategies have been utilized to solve it in the past, and finally the current and future plans of action to tackle this problem. Is the West the main responsible entity for the solutions formulation, such as shaping water-related policies and providing technocratic solutions as well as its consequent implementation, or is the weight more on local actors? These are some of the questions that are key for this frame analysis and which I use to critically scrutinize the framings under study. Therefore, the academic relevance of this research lies in the fact that it is looking at the problem solely from a social science aspect whereas the big bulk of the research published to date is more focused on defining the causes and solutions of the “problem”. My field study took place in Maputo, the capital and political, economic and demographic core of Mozambique, and where most of the actors involved in the water sector are located. The research question is as follows:

How is the problem of urban water scarcity framed by different actors in Maputo, Mozambique, and how does this shape the solutions proposed?

Sub-questions are as follows:

1) Which actors/groups/stakeholders are involved in the problem framing of Maputo’s water crisis? How do they differ in social power?

2) How do they frame the problem differently and on which basis is each framing justified?

3) What types of solutions (techniques and/or policies) follow from the different problem framings and who is held accountable for formulating these solutions and implementing them?

4) How do the identified framings relate to one another and which drivers are behind the dominance of one framing over the other?

(11)

3

2. Chapter Two: Theoretical Framework &

Operationalisation

Water scarcity is often difficult to define as a supply or demand problem. Opinions differ in labelling it as scarcity in the physical sense, or as a result of weak resource management and a geographically and historically situated social fact. In other words scarcity can be defined as a socially constructed phenomenon (Arnaud, 2010). Taking Trottier’s work as an example, I base one of my assumptions on the fact that water scarcity is not merely a “natural” phenomenon, therefore conducting a critical analysis of water scarcity and accepting the subjectivity of the term. Its causes have different understandings by different stakeholders, leading to a variety of solutions (Trottier, 2008). For example, this can result in some solutions being technocratic, relying solely on technological innovations, thus possibly embedded in a framing which locates the cause of water scarcity in natural processes.

Based on various reports1 and media articles2, water scarcity is at the forefront of this water problem. Although this study does not aim to measure water scarcity in Maputo, I do look at the definition of the term as I expect to find it, in one way or another, manifested in the framings I identify through my data collection. Over the past two decades, water scarcity has had many different definitions based on the different methodologies employed to measure it, all using unique sets of indicators and considerations. These include, but are not limited to: basic human needs, population growth rate, ecological water requirements, industrial water usage and the level of exploitation of freshwater. The most widely known is the Falkenmark Water Stress Indicator that was developed in 1989 and set the groundwork on which many other methods were later formulated (Brown & Matlock, 2011). As is the case with several indicators, it measures the relationship between water availability and human population, and leaves out the more complex aforementioned social indicators, as they have proven to be difficult to measure. It defines as “water scarce” countries where the measurement of freshwater per capita is under 1000 m3 (Rijsberman, 2004). According to this water stress indicator, 62 per cent of the world population will experience water scarcity in 2030,

1_{Example: UNICEF Drought Humanitarian Situation Report:}

2_{Examples include: Article 1:}_{https://allafrica.com/stories/201802210911.html}

Article 2: https://www.theguardian.com/environment/2018/apr/22/mozambique-prays-for-rain-water-shortages-hit-poor-maputo-southern-africa-drought

(12)

4 Mozambique being within the group of countries that falls in the category of “under 1000 m3” (Rijsberman, 2004).

The aim of this research is to take a deep look into the water scarcity problem in Maputo and examine the different perspectives relating to it. I conduct this by means of a type of discourse analysis called frame analysis. As described by Kitzinger, “Framing refers to the process whereby we organize reality – categorizing events in particular ways, paying attention to some aspects rather than others, deciding what an experience or event means or how it came about” (Kitzinger, J. 2007). This type of discourse, as conceptualized by Hajer, revolves around “an ensemble of ideas, concepts and categorizations through which meaning

is given to phenomena” (Hajer 1995). I examine the problem through the different lenses of the stakeholders involved, defined by Bacchi (2009) as the “problematisation.” I conduct an analysis to understand how the discourse surrounding water scarcity shapes its reality, giving equal weight and consideration to the different points of view on the issue. I identify the problem framings constructed by all actors involved through an objective lens, without labelling them as “right” or “wrong”. Similarly to the study by Cairns & Stirling on the framing of climate geoengineering, I do not aim to “define” water scarcity; rather, I treat it as a “discursive phenomenon, the bounds of which are continually being negotiated” (Cairns & Stirling, 2014). As per Goffman (1974), the framings being analysed represent “schemata of interpretation” which contribute to the shaping of the identity of certain phenomena and the consequent events related to it (Benford & Snow, 2000). This encompasses a variety of research practices with different levels of emphasis on particular aspects. However, all mentioned studies reject the notion of an objective truth and consider knowledge as disputable. It is crucial to take into account the practices and the dominant language used to describe the problem by different players and their effects in shaping power and knowledge structures (Feindt & Oels, 2005). “Knowledge” related to the issue is embedded in the different problem framings. Each point of view has the potential to present a new perspective of the issue, consequently shaping a basis allowing stakeholders to endorse a particular solution (Feindt & Oels, 2005).

The research posits that the different framings around water scarcity in Maputo are the result of existing power relations. Each of these framings purposely uses or discards knowledge, serving specific interests. From a Foucaultian point of view, “Power is everywhere; not because it embraces everything, but because it comes from everywhere” (Foucault, 1998, p. 93). Power dynamics that shake the state of affairs are often disguised by the idea of a single rationality. Through the power structures in place, stakeholders are able to

(13)

5 stress on particular aspects of the issue, giving it more weight than the others, and therefore the language and labels utilized serve to directly impact the general circumstances. Hence, “framings in and of themselves can be understood to perform particular functions” (Cairns & Stirling, 2014). Therefore, the research will be based on a social constructivist perspective of the issue, rejecting the idea of a single interpretation of reality.

The frame analysis will be used to critically look at “apolitical pieces of knowledge”, potentially exposing the underlying political accounts at play (Feindt & Oels, 2005). In their study on the public opinion on genetically modified (GM) crops technology, Mayer and Stirling (2004) found that all actors involved in the issue had their own motives; private companies, media outlets, government bodies and nongovernmental organizations alike. They especially call into question the scientific evaluation of the situation as it is often based on assumptions that are biased: “In the end, the business of posing questions and shaping assumptions is as much about politics as it is about science” (Mayer & Stirling, 2004).

A thorough literature review prior to the fieldwork illustrated that in the case of Maputo, water scarcity is the most widely used framing and is affects different groups in a variety of ways. In order to operationalise the theories mentioned above, I follow Bacchi’s strategy for analysing policy, which is through “problematisation”. This concept is based on “politicizing taken-for-granted truths” (Bacchi, 2012). As manifested in the operationalisation table in the next section, it is constructed based on the following steps.

1. Uncovering the problem being contested, in this case water scarcity. And

looking at who is framing this problem.

2. Finding out how the problem is represented as well as the justifications for specific representations, thereby tackling the “how” and “why” questions. This aims to uncover the assumptions, knowledge, and/or concepts used to justify the framing. This is a type of “Foucauldian archaeology”, which aims at revealing political logicalities leading to each representation.

3. Analysing the consequences that emerged from each framing. This is both

in terms of which effects have already been impacting populations, what is being planned for the future, and how will these consequences affect the status quo. Bacchi illustrates this in the following question: “How does the attribution of responsibility for the ‘problem’ affect those targeted and the perceptions of the rest of the community about who is to ‘blame’?” (Bacchi, 2009)

(14)

6 In my analysis, I look closely at these representations in order to find out the intentions and/or origins of each, questioning the purpose they are trying to serve. And based on that, I draw some comparisons between five framings, identifying the aspects in which they overlap and differ, through the identification and analysis of these differing problem framings.

2.1 Conceptual Scheme

The conceptual scheme below illustrates how the “problematisation” of the water crisis takes place. As demonstrated in the theoretical framework, there is not “one truth” in the politicized environment. Therefore, various points of view on the water problem exist in the black rectangle titled “the politicized environment”.

The blue oval is the lens of perception that creates the different framings, by incorporating three different categories, those of knowledge, actors and solutions. Each of these categories is determined by different factors listed on the far right; they represent the key information I seek through my data collection.

Figure 1: Framings of the water problem: conceptual scheme

(15)

7

2.2 Operationalisation Table

Dimensions Variables Indicators

Problematisation

Actor

- Who is doing this framing? - What is their influence/decision

making/governance level?

- Who are considered to be the injured parties?

Knowledge

- What are the types of evidence utilised for problem identification?

- What is the justification for using such information? Is the specific information used to serve certain interests?

Problem

- What is considered to be the ‘problem’ in the discourse?

- Why is this considered to be a problem? - Which assumptions underlie this problem

framing?

- Which ideological and political beliefs are grounding the problematisation?

- Which are the key concepts, binaries, and categories used in the discourse?

- What kind of problem is it made out to be?

Solutions

Solution

- What solutions (techniques and/or policies) are mentioned?

- What is the rationale behind the proposed solutions?

- What has been done in the past and what impact did it have?

Responsible actors (Including decision making)

- Who is held responsible for the

implementation of these solutions? And ultimately, who is accountable?

- Is decision-making a participatory process?

- What is the source of funding?

Table 1: Operationalisation Table

(16)

8

3. Chapter Three: Methods

In this chapter, I explain in detail the methodological scheme I used in Maputo. Based on the conceptual scheme and operationalisation table presented in the previous chapter, I unpack the data gathering process including sampling and analysis methods. I then reflect on the ethical aspect of my research including its overall limitations.

3.1 Data collection

My analysis is based on qualitative data collected by means of in-depth key informant interviews supplemented by a review of key stakeholder literature. This approach presented me with a critical lens through which to view the complicated relations between the stakeholders and how it shaped the status quo. The utilised methods are expanded upon below.

For the purpose of data collection, I spent eight weeks in Maputo where Dr. Adriano Mateus Biza acted as my local supervisor. He is part of the Department of Integrated Water Systems and Governance at the IHE and works on the UDW Maputo project as a PhD candidate. His project aims to explore the way wastewater is governed in Maputo city, explain the source of sanitary divide in the city and the ways in which people’s relations with wastewater are shaped by policies framings, rationalities, politics and varied spatial and even temporal

processes3_{. My first meeting with Adriano was at the Eduardo Mondlane University (UEM)}

where he gave me a general overview of the structure of the water sector and provided me with attendance lists of water-related events in order to find contacts and start scheduling my interviews. Therefore, my participants were selected based on their position within the Mozambican water sector. Gradually, I started interviewing some respondents who in turn led me to other relevant contacts, using snowball sampling, a feasible way to proceed in such a context.According to Hennink et al. (2011), this non-probability method’s limitation is that it can result in research participants who are all within the same social network. However, in this particular case, this does not form an obstacle, as the opinions that I was gathering were those of different stakeholders within the water sector, which in itself forms a professional network. The actors within the water sector were not only able to lead me to their professional counterparts from different organisations, but also to individuals at a higher level that they know or are acquainted with through networking events or personal relations. Platforms in the water sector often include — or aim to include — ministry representatives as

(17)

9 well as international and national NGOs with the aim of sharing different points of view and planned and on-going initiatives.

3.1.1 Document Analysis

Reviewing existing documents helped me understand the approaches of the agencies I planned on interviewing. Ministries, donors and aid organisations produce reports related to their water strategy, which helped me supplement the data collection I later conducted through interviews. Therefore, I did not rely on the information attained from these documents to introduce the different framings. The points of view derived form this document review served solely as a basis for data collection; unlike the data retrieved from the interviews, they were not treated as raw data and thus were not coded in Atlas.ti. Whenever I came across valuable information, I documented it and looked for cases where opinions from different documents may be similar or where they vary. In some instances, these documents helped me obtain the contact information of certain actors.

I conducted this secondary data analysis prior to traveling to the field, during the fieldwork period and upon my return. I had originally planned to finalize the analysis of all available documentation before starting the data collection in order to go in the field with a better understanding of the context as I saw a considerable advantage in the feasibility of such a method in terms of timeliness and efficiency. However, I found that some of the information did not end up fitting any of the framings identified through the interviews. This was not in relation to the whole aspect of a framing; rather, it is mainly related to some new technologies being researched, tested or implemented. In the majority of the cases, towards the end of the interviews, I found myself bringing up such solutions in my questions, as they did not come up in the direct answers of the respondents. While I do acknowledge that all reports belong to a particular framing, I am also aware of the scale of my study (data collected and organized into five framings), and that not all collected data will necessarily fit in its scheme. It did, however, make me appear more knowledgeable about the subject in front of my interviewees and therefore more confident. I later realized that some of the documents I spent a significant amount on were not very relevant to the focus of the research. It was not until after finalizing the first few interviews that I became selective in my choice of documents to examine and did so in a much more efficient way, adopting an iterative process. However, this also meant that the first few interviews could have potentially included more insightful questions on my part.

(18)

10 I found a particular benefit of doing the document examination at these different stages; firstly, because that gave me a general overview of the different activities conducted by different stakeholders, which greatly shaped the type of questions I was prepared to ask in certain interviews. Upon my return to The Netherlands, conducting additional secondary data analysis proved to be a lot more efficient as I already had a good grasp of the situation, allowing for the analysis to run more smoothly than at the beginning and allowed me to easily triangulate the information. However, I still detected some nuances that could have been interesting to inquire about during my interviews and some stakeholders that would have added more credibility to the study, but that is a limitation that is out of my control at this stage and that is unpacked in “section 6.4: Research Limitations”.

The documents analysed consisted of scientific literature mainly obtained from universities, media products including online news sources and local papers, civil society groups and Non-Governmental Organisations’ reports as well as government and donor reports. Inherently, the use of secondary analysis poses the risk of the lack of control over the quality of the data received. As expected, many state-produced publications were significantly out of date, therefore, the quality of data was not always up to the standard, forcing me to use a critical lens throughout and consult my local supervisor at times. The criteria I used to select the documents to be analysed was firstly the relevance of the topic and then the relevance of the owner of the document within the water sector.

3.1.2 In-Depth Interviews

In order to identify the different framings, I conducted 22 interviews, tackling all the sub-questions (p.2) as well as indicators found in my operationalization table (p.7) with each of the interviewed stakeholders. These interviews took place in Maputo with different actors from the local community such as government officials, civil society, non-governmental agencies and private entities as well as researchers. For each interview, I start by presenting myself and then ask the respondent to describe their role within the organization of which they are a part. After that, I continue by asking them whether there is a water problem in Mozambique and if so, how do they define it and how do they experience it. Time spent with each participant depended on his or her schedule and availability, which was often but not always in correlation with their position in the institution; the higher the authority level, the less time was available for them to engage in the study. Interviews lasted between 45 minutes to 2.5 hours. I did not use a defined interview protocol with prepared questions. Rather, I grouped a list of themes that I wrote in my notebook and proceeded with the interviews. In

(19)

11 the instances where I felt that the conversation was moving too much in a different direction, I went back to my list to ensure that all subjects were covered.

The main challenge I was faced with was the language barrier. Language is a very important element of any type of discourse analysis. Mozambique’s official language is Portuguese, with several other indigenous languages, Makhuwa being the most dominant Bantu language. As I do not speak Portuguese, I sought the assistance of a university student for translation. Unexpectedly, the majority of the interviewees insisted on conducting the interview in English. I presume that the presence of the translator with me showed my preparedness and willingness to do the interview in Portuguese, which helped me gain interviewees’ respect for going the extra mile.

3.1.3 Informal Conversations

An ex-colleague of mine currently works for a nongovernmental agency in Maputo and she served as a useful link for me to meet new individuals in the development sector. This broadened the spectrum and type of persons I engaged with in informal conversations and allowed me to listen to the opinions of development workers in different sectors, other than water. At the end of each day, I took it as a habit to write down the highlights in my journal. Due to the absence of barriers during these conversations, some opinions were more “daring” and fit well with the framing n5, that of corruption.

3.2 Methods of Analysis

Frame analysis was the main tool manifested through the use of this research method. The combination of these methods allowed me to compare the secondary data analysed with the different points of view captured through the interviews, which in turn enabled me to get a clear idea of people’s opinions through the essence of their own words. As foreseen, this process further exposed the socially constructed dimension of water scarcity by allowing for certain groups to voice their opinions. Considering the limitations of such a method, information received from interviews is often subject of bias due to the nature of the interface.

Problem frame is the unit of analysis of this research. The frame analysis requires the transcription of the interviews but this does not necessarily have to be a full transcription, as long as it captures the essence of the opinions being voiced. I conducted the transcriptions partly in the field and the rest when I returned. I based the selection of which interviews to prioritize for transcription on the importance of the interviews; some included key data and

(20)

12 figures that I needed to dig deeper into before going for another interview the next day, while others were not associated with that level of importance so I could afford to do them at a later stage.

Regarding the organisation of my data, I started by doing it manually based on the themes I had highlighted while transcribing. However, after using Atlas.ti as a trial, I found that to be much more efficient and I proceeded accordingly. The indicators in the operationalisation table served as a guide for the chosen codes and sub-codes. Thus, I assigned codes to each of the different themes; these consisted of the main reasons for the problem and the main solutions, divided in short, medium and long term. Additionally, I selected the most significant passages to include as quotations in the thesis.

After finishing this exercise, the next step was to identify the prevailing framings and then study the relationship between them: their interrelationships, the aspects in which they differ and ultimately having clearly identified groups of framings. I studies the validity of the results obtained through triangulation of the different qualitative methods used, aiming to reduce the bias that could result from using solely one method. This also depended on the quantity of data that was gathered throughout the fieldwork, as the increased number of perspectives documented allows for better interpretation through triangulation (Thurmond 2001). The amount of data collected through the interviews adequately fulfilled that requirement.

Another issue that was made visible to me upon my return from the field was the structuring of the interviews. Initially, I had prepared an interview guide based on my operationalization table, which I reviewed together with my local supervisor. I used it as a guide for the first interview, after which I realized that giving more freedom to respondents to express themselves about what they feel is most important at that moment would provide me with better quality data. This was because I sometimes felt that I was steering the conversation in a clearly drawn direction that had started to become redundant. I found that giving extra freedom to respondents was useful for getting more information in a short period of time, however, this had made the data analysis – coding – much more difficult and time-consuming.

(21)

13

3.3 Ethical Considerations

The focus of this research is on problem framing, which is a form of discourse analysis, and the topic being framed – the Maputo water problem – is not connected to the respondents’ personal and sensitive issues. Thus, there is no substantial potential harm, discomfort, physical or psychological risk in sight that could relate to personal or traumatic experiences. Nonetheless, the water problem, or water scarcity per se, is a contested subject in Mozambique. When it comes to how it is experienced by local communities, it can also be a sensitive issue due its political aspect. A key consideration I took into account prior to my departure to the field is that in some situations, some participants may be reluctant to share information with me regarding their own strategies or the strategies of other actors.

However, this consideration only materialized with one interviewee when it concerned government documents and the person did not feel that they were in a position to share those documents with me. Instead, they preferred that I try and obtain these documents myself from the government institution. My request attempt was not met with a positive response. While I did not ultimately obtain the documents, my conversation with the government entity was enough for me to get the information needed.

Conducting my study as part of an IHE Water Institute project in association with the Eduardo Mondlane University (UEM) in Maputo has opened doors for me. With the help of my local supervisor, I was able to obtain a signed letter of Introduction4_{from the head of the} Anthropology & Archaeology Department at the UEM stating that this research is part of the requirements to complete the Master degree in International Development Studies at the University of Amsterdam. It also clearly mentioned the aim of the research, emphasizing that analysis would be done giving equal weight and consideration to the different points of view on the issue. The letter of introduction was a requirement for some government agencies, mainly the semi-private ones, to give me permission to do an interview. Most of the other agencies did not ask for any written documents. I felt the need to explain in person, not only as a good practice, but also to emphasize that, although I am a student from a Dutch university, I do not represent the opinions of the Dutch embassy – as it is a very important actor in the Mozambican water sector – and that the aim of my work is not to promote an ideology or a specific solution. This helped me gain the trust of my interviewees, allowing enough space for them to be critical about the projects under discussion. Thus, all participants were informed of the purpose of my research and how the results will be dealt with before

4_{A copy can be found in Appendix}

(22)

14 they gave their consent. Prior to each interview I obtained verbal informed consent of the participants, ensuring that participants had adequate information about the aim of the study. Confidentiality and anonymity were proposed from my side, especially due to the political nature of the topic and the power dynamics at play, however, none of the respondents opted for that as they felt comfortable enough discussing the topic openly. In one instance, I was asked by the interviewee to stop recording the meeting, as the person felt uncomfortable criticising government institutions on record. Therefore, I respected the person’s anonymity, reflecting my commitment to ethical procedures. Finally, the findings will be shared with the researchers from the IHE and UEM as well as the participants who expressed their interest in knowing the study findings.

(23)

15

4. Chapter Four: Contextual Background

4.1 Mozambique: General Information

Mozambique gained its independence from the Portuguese Empire in 1975, after which a 16-year long civil war took place and ended in 1992. This has led to a big presence of international actors in country, being strongly involved in alleviating the situation of refugees through the war and “reconstruct the country” after it. This is mainly done through the work of big organisations such as USAID, World Bank and various UN agencies. They all implemented programs aiming at reducing poverty through boosting agricultural production, awareness on health (mainly focusing on HIV and Malaria), nutrition and sanitation as well as increasing access to clear water (The Borgen Project, 20175). These projects, while alleviating the poverty of some of the most marginalised communities, leaves many Mozambicans still struggling to survive. According to the World Bank, “approximately 17 per cent of under-five deaths in Mozambique are the result of diarrheal diseases, primarily caused by poor water and sanitation,” (World Bank, 20136).

Economic growth in Mozambique has been unsuccessful in ending poverty. More specifically, the problem lies in the fact that the growth has not reached Mozambicans equally, perpetuating poverty effects further. Estimates published by the World Bank reveal that “more than 2 million additional people could have been lifted out of poverty had Mozambique’s economic growth between 1997 and 2009 been more equally shared” (World Bank, 20167). This vast inequality confirms the aforementioned8 concept of the social construction of water scarcity, as with economic growth, individuals belonging to different social groups experience it differently.

Weather seems to be the most common reason for hindering poverty reduction, as stated in most international reports and media articles. Therefore, many efforts are geared towards the prevention and mitigation of extreme weather conditions. “Three in four farmers in Mozambique report losing part of their crops, animals or other equipment due to climatic shocks. A child born in a place affected by severe floods, like the ones that hit the country in

5_{Retrieved on 17 January 2018 from:}_{https://borgenproject.org/about-poverty-in-mozambique/}

6_{Retrieved on 17 January 2018 from:}

http://www.worldbank.org/en/news/press-release/2013/07/25/mozambique-world-bank-clean-water-urban-residents

http://www.worldbank.org/en/news/opinion/2016/12/21/picking-up-the-pace-of-poverty-reduction-in-mozambique

(24)

16 2000, is more likely to be undernourished, drop out of school, and not participate in the labour market later in life than a child raised under normal weather conditions”(World Bank, 20139).

The discovery of a large reserve of Liquefied Natural Gas in Northern Mozambique came with immense promises in terms of job creation and lifting the population out of poverty. However, so far, this has solely benefitted the elite. Another major recent event affecting the economy is the April 2016 scandal that uncovered over USD 1.4 billion in hidden debt from the IMF, leading big donors to suspend their aid to the country, decreasing the flow of capital and highlighting Mozambique’s dependence on foreign investment. Simultaneously, Mozambique is experiencing one of the highest population growth rates in Africa, with an increase from 2.4 per cent to 3.5 per cent in 10 years time, keeping in mind that the sub-Saharan African average is 2.7 per cent per year (Hanlon, 201810).

4.2 Maputo: Research Location

Mozambique’s capital, Maputo, is located on the southern coast of the country. During the aforementioned civil war, the city experienced rapid urban growth as families fled the violent conflict in the rural countryside and looked for safety and a better livelihood. In recent times, residential and industrial development has spread to Maputo’s neighbouring city and district of Matola, Boane and Marracuene creating a Greater Maputo metropolitan area. However, unlike the old city centre of Maputo (known as The Cement City), Greater Maputo does not benefit from centralized drinking water supply nor from a wastewater network (SEED, 2010). As for Maputo’s climate, it is tropical and more arid compared to the Northern part of Mozambique. The summer season, which is the wet season, extends from December to March with most of the precipitation taking place in the month of January (rainfall average of 160mm). During this time, the city is susceptible to flooding, which has caused a significant amount of damage in the past. Winter, which extends from May to September, brings with it very little rainfall with August being the driest month (rainfall average of 15mm11). The latter period puts great pressure on the Cement City’s water resources, as it depends primarily on freshwater, mainly on the water coming from the Pequenos Libombos dam.

http://www.worldbank.org/en/news/press-release/2013/07/25/mozambique-world-bank-clean-water-urban-residents

10

http://www.open.ac.uk/technology/mozambique/sites/www.open.ac.uk.technology.mozambique/files/files/Mozambique_400 -10Jan2018_Census_Climate-change_books.pdf

(25)

17

4.3 Water Governance in Mozambique: General

In this section, I explain the roles of the different actors in the water sector, their decision-making level and roles and responsibilities as described throughout the interviews. It is important to note that most respondents called attention to the complexity of the Mozambican institutional framework. I start with the most influential directorates, which are both under the Ministry of Public Works, Housing and Water Resources: DNAAS (the National Directorate of Water Supply and Sanitation) and DNGRH (National Directorate of Water Resource Management). Until 2014, the main two institutions – DNAAS and DNGRH – formed a unit board under the name DNA (National Directorate of Water) before they were separated into these individual entities in 2015. After that, I give an overview on the private sector participation, the donors and finally, the synergy between the water and sanitation sectors.

4.3.1 Public and Semi-Private Entities

DNAAS is responsible of the provision of water supply and sanitation in the country. This

includes both coordination and promotion of infrastructure for water supply and sanitation. They report on national water and sanitation in the country (urban and rural) to the Ministry of Public Works, Housing and Water Resources. DNAAS is responsible of the water network, which is the focus of their priority investments; this includes the sources, the piping system technicalities, storage, and tendering of private companies for implementation. In their words, they negotiate the “big ideas”, including national plans and international agreements for interventions. DNAAS works with two major agencies: FIPAG, which is concentrated on urban water and AIAS, which deals more with sanitation. DNAAS puts in place the standards for the water network system. After that, it is FIPAG that contracts private companies for the system’s construction, ensuring that they follow the standards set by DNAAS. Thus, FIPAG handles the direct implementation in water supply while DNAAS oversees the situation and takes priority decisions related to the size and type of investments and gives guidance on methodology and regulations. AIAS and FIPAG are on the operational side.

DNGRH is the entity responsible for water resource management. Their role consists of drafting policies, water laws, flood and drought management plans, and the promotion and

(26)

18 construction of water infrastructure (construction of dams, excavation of reservoirs etc). They are also in charge of the national hydrological monitoring; the monitoring of water levels and rainfall patterns etc. They operate at the national level, while ARA (The Regional Administration of Water) conducts the data collection and reports to them on all the different regions of Mozambique, in order to feed into the national database. So, ARA, the regional water authority, is responsible of the source. In the case of Maputo, situated in the southern part of the country, it falls under ARA Sul (ARA South). Apart from reporting on the hydrological situation of the different water sources and advising on the flow that should go through the dam to ensure acceptable levels, ARAS are in charge of selling the raw water to different clients, the biggest being FIPAG – a water asset publicly owned company that is the “owner” of the network – for domestic supply. From the intake, FIPAG is responsible for treating the water and finalizing the network connections in the major Mozambican cities. FIPAG and ARA Sul both operate under the same ministry, with FIPAG being more independent in terms of administration and finance.  FIPAG manages one big system in Maputo and smaller systems outside Maputo. They have sub-contracts in place with different private entities for the management of the network, the biggest being Adem (Aguas de Regiao de Maputo), a private company in charge of the operations of the entire region and in which FIPAG is a shareholder. Within the service area of Adem, there are some areas not covered by the main system, as the FIPAG coverage is roughly 50 per cent. Therefore, these areas are covered by small private systems, typically from groundwater. Other than having contracts with these big and smaller entities and monitoring their performance through regular monitoring, FIPAG also provides them with technical assistance.

AIAS (Infraestruturas de Abastecimento de Água e Saneamento) manages the infrastructure for water in small towns, and sanitation for the whole country. This consists of setting up the drainage and sanitation plans. While they are also under DNAAS, they have minimal coordination with FIPAG and operate using their own funding, having full institutional autonomy. They complete the rehabilitation and expansion works and then set up an agreement with private sector entities for their management. Unlike in the case of sanitation, the municipalities rarely manage the water infrastructure; rather, they act as beneficiaries. According to AIAS, the municipalities do not yet have the capacity to handle this kind of process, as they do not exert a strong authority on the communities that would ensure the payment of the latter for such services. It seems that when the municipality is in

(27)

19 charge, people feel like the service delivered is more like a given – a right. This is why the private sector oversees this task.

CRA is the water regulatory board in Mozambique that is responsible for “ensuring the balance between the quality of water supply service and the provided sanitation, the interests of users, and the economic sustainability of water supply systems, through the exercise of economic regulation” (WashWATCH, 2018). CRA’s major role is the setting of water tariffs and the regulation of water supply standards. For example, in order for the state-owned companies providing water (Adem) to survive, they must be able to cover all the costs related to water provision and ensure recovery on their investments as a company with the help of these tariffs. However, it is CRA that decides on the final figures. If the tariff is too high, CRA handles the complaints regarding it and works on regulating it by studying the situation, taking into account both the service providers and the community members and using specific indicators as basis. The tariff would then need to be approved by the Council of Ministers. Being an independent government body, CRA report directly to the Prime Minister.

INGC is the National Disasters Management Institute, which is an autonomous institution under the Ministry of Foreign Affairs and Cooperation. As the name suggests, it handles disaster management affairs. Within the frame of this study, it is important to be aware of INGC’s role as the entity that ultimately activates different alerts for the country. This happens after the relevant government bodies falling under the Ministry of Public Works, Housing and Water Resources present to INGC the technical data in the presence of the Council of Ministers. A green code refers to the normal situation during which development projects are implemented coupled with on-going monitoring. An orange alert –

which is the one that was activated for Maputo on 13 February 201812 – is activated when

there is an imminent event that might impact the population and have humanitarian consequences. In such a situation, the government should have a plan to respond to the situation and needs to conduct daily monitoring of the phenomena and report it to INGC and the Council of Ministers. The highest level is the red alert, which is activated when the emergency situation has already impacted the population and international help is sought. This is evidently the most difficult alert to activate, due to the political implications as it declares that the country is in a state of emergency.

(28)

20

4.3.2 The Private Sector:

The endorsement of the National Water Policy in 1995 is a key event for the Mozambican water sector. This opened the door for increased implementation by the private sector in urban water supply and decreased the government presence in the operations side of water and sanitation services. This also included the acknowledgment of water and sanitation services as “both social and economic goods13_”.

“Around ten years ago, when FIPAG finished implementing their project, there was the issue that the private sector should stop supplying water so that the state can start assuming that role. That is not wrong, because I think that providing water is the role of the state in any nation, but what happened is that the state entities lack the capacity to supply water for all the city of Maputo, making the role of the private sector essential.” (Participant #1)

When discussing the private sector involvement in water supply, different points of view emerge. Some view it as a symbiotic relationship between the government and the private sector, while others see it as a direct consequence of the failure of the state to go through with its plan of supplying water to the whole population. The most prevalent opinion is that the state is making efforts to regulate these commercial entities – not because of the water crisis or lack of surface water per se – but rather because of their increasing number and in order to have some sort of control over them. This led to the creation of a framework for delegated management within the water sector. In the water supply subsector, there is FIPAG (the development fund for water supply) and the private operators. There is a delegated management of water allowing for the private sector to operate state assets; and when the right to operate in a certain area is given to a private operator, a form of exclusivity is given to them over that specific area. Thus, private operators are spread across different areas in the city and operate legally within concession areas.

Initially, the PoPs (Pequenos Operadores Privados14) were called “informal service providers” but nowadays most of them are recognized as “formal service providers”. In order to obtain this status, they must first apply for a license from the municipality declaring the economical nature of the activity and pay a fee to ARA Sul (indirectly to DNGRH, the entity managing groundwater resources) in order to obtain an authorization in the form of water

13_{http://wsp.org/sites/wsp.org/files/publications/CSO-Mozambique.pdf} 14_{Portuguese for: small private operators}

(29)

21 permits allowing them to construct boreholes. In return, they must comply with certain obligations related to quality standards and service hours. However, these are unclear to most interviewees. There is a general understanding that some kind of regulation should be there to ensure the sustainability of groundwater, but all entities related to groundwater acknowledge that the capacity is not there yet to ensure the sustainable exploitation of this resource and that there is an overall lack of surveillance and onsite monitoring.

“ARA Sul also became responsible for groundwater four years ago, it’s still a new process. We have more time and experience managing surface water. People are not very familiar with the groundwater harvesting process.” (Participant #18)

Back in 2012, the Ministry of Public Works, Housing and Water Resources created a tool to

regulate the exploitation of groundwater. Today, this still serves as a guide for ARA Sul; it mainly includes the rules of drilling and extracting groundwater as well as its distribution and the method used to charge it. When it comes to groundwater monitoring, the process is very expensive on their part, which led to reducing its frequency from monthly checks to every three months. In order to conduct this monitoring, a contribution is needed from the entities exploiting it.

Agreements are established between the PoPs and the municipalities; however, the latter have minimal involvement in the management of water supply in general. Thus, with regards to accountability, PoPs deal directly with ARA Sul. Moreover, to date, many of the relations with other entities are not well established through clear mechanisms. For example, with CRA, the regulatory body, the private sector relationship is starting now and the level of engagement is limited to the tariff updating process.

“But even years ago, prior to the drought scenario, the scenario of great population growth was in the position that all the others were intruders in the water business, in water exploration. While in the first approaches the government said not to exploit groundwater that could be depleted, the government appears to adopt our systems, the government is already opening boreholes like us. Now the government views private operators as partners in the process of solving water scarcity problem in urban and other centres” (Participant #6)

(30)

22 guaranteed that it is around double that of the public one. This is mainly because the private systems lack water meters and use fixed rates instead, regardless of the amount of water used by the consumer. However, according to an AFORAMO employee, the difference is insignificant15 and is justified by the obligation of the private sector to pay taxes.

“If I had the possibility of levying tax, I would reduce the price of water” (Participant #7)

In terms of service hours, the private systems have gained the trust of their users as they can consistently supply 24/7 and remain unaffected during droughts. In terms of water quality, no precise data could be found for both sides. Finally, regarding coverage, the main network does not pass through the suburban areas, thus they are all covered by the PoPs. The exact percentages of coverage of public vs. private were not mentioned by any interviewee and could not be found on any public platform. Some interviewees believed that private providers cover the majority of Maputo, whereas an Adem employee affirmed that they cover 66 per cent of the city, indicating that the rest (44 per cent) is privately serviced.

AFORAMO is a non-profit association consisting of the heads of local private operators (PoPs). The association works as an interlocutor representing the members’ interests and voicing their concerns. In order to become a member, associates have certain obligations to fulfil, one of them being a monetary contribution. Members are expected to adhere with the regulations, as stated in their mandate. The association does not produce anything per se, meaning that all fees go straight for the survival of the institution.

4.3.3 Donor Agencies:

There is a strong flow of institutional and bilateral aid channels throughout the country, although the latter has seen a decrease since the hidden debt scandal in 2016. Embassies play an important role in funding for water initiatives as well as chairing coordination platforms, such as the Dutch Embassy with the GAS (Grupo de Água e Saneamento), which provides a multi-stakeholder coordination platform for Integrated Water Resource Management. It involves donors, the government institutions mentioned in the previous section16 and various national and international agencies and implementing partners specialised in water. Participants include: the World Bank, African Development Bank, DFATD, DFID, UNICEF, WSUP, the Embassy of Japan, JICA, Wateraid, USAID, and Oxfam, among others.

15_{It is about 10 to 20 Meticais (as per the interview with an employee from AFORAMO)} 16_{“Public and semi-private entities” section 4.3.1}

(31)

23 Embassies often work by providing capacity to the government, but especially since the hidden debt scandal, funding is not channelled directly through the central government and implementation is done by private or semi-public entities.

4.3.4 Governance of Water & Sanitation:

“Mozambique did not meet the MDG’s target for Water Supply and Sanitation. The national access level as per Government of Mozambique (GoM) reports stands at 42.2 per cent for rural water supply and 20.7 per cent for sanitation by end of 2016. Access level in 2010 as per the JMP stood at 27 per cent for rural water supply and 13 per cent for rural sanitation17.” (AFDB, 2017)

As mentioned in the previous section, water and sanitation are under the same ministry. When it comes to infrastructure, it falls under AIAS and the operation of sanitation is with the municipality. The municipality has much more involvement in sanitation than water. In Maputo, the public operator is linked to the water sector for the water supply, the central government being highly involved. The same does not apply for the drainage system, as the role is given to the municipality. All interviewees confirmed the dislocation between the water and sanitation schemes of the country and the difference in investments between them, with sanitation almost never seen at the same level of importance.

“Water and sanitation schemes are completely two separate things. They belong to the same ministry; they even share offices in some places. But they are going in completely different directions, without any coordination between the two. Investment in sanitation is much lower than investment in water. People are more willing to pay for water than sanitation” (Participant #4)

The sanitation network in Maputo works through small collections of money and donations, as it is not yet part of the water tariff. The idea of adding a sanitation fee to the water tariff is a municipality initiative that has been under discussion for around five years. The sewer system currently only covers a very small part of the city (around 10 per cent18). Thus, the rest of the population (around 90 per cent) that is not connected is the portion paying for sewage services such as trucking services, and then the trucking companies pay a fee in order to discharge the sewage at the wastewater treatment plant site. The estimated cost for the rehabilitation of the current wastewater treatment plant is USD 22 million with no

17

https://www.afdb.org/fileadmin/uploads/afdb/Documents/Project-and-Operations/MOZAMBIQUE_- _National_Rural_Water_Supply_and_Sanitation_Program__PRONASAR__in_the_Nampula_and_Zambezia_Provinces_-_PCR.pdf

(32)

24 funds in sight to tackle this issue, seeing as the water problem remains the priority.

“The big dream is to have a wastewater treatment plant but it is a very big thing and it is very hard to find funding for such a thing. It is much easier to obtain funding for the provision of water, unfortunately. But it seems that times are changing, and sanitation will start playing a bigger role” (Participant #3)

Up until 2015, water supply and sanitation for both urban and rural areas formed one department. After that, a restructure took place and the following makeup was created: one department that deals with water for both urban and rural areas, and another one dealing with sanitation for urban and rural. The rationale behind the restructure was that the water company was growing at a faster pace with higher percentage coverage in rural areas, unlike the case of sanitation. As per DNAAS, the expectation was that the separation of these two entities would lead to more attention on these issues, allowing sanitation to catch up with water.

“But still, we believe that water and sanitation must go hand in hand, that’s why we had the PRONASAR program”(Participant #2).

PRONASAR is the National Rural Water Supply and Sanitation Program in Nampula and Zambézia provinces, implemented in the period of February 2012 till September 2017 and funded by the African Development Bank. However, similar initiatives were not applied in urban areas.

The NGOs interviewed all shared the same opinion regarding the importance of the integration of water with sanitation schemes. At the design stage of their projects, they try to ensure as much as possible an integrated approach that includes the components of water supply, sanitation and hygiene with the latter mainly focused on behavioural change and water conservation. The overall objectives cannot be achieved if water is provided but poor hygiene persists. However, this depends on donor strategies for both sectors.

“Water and sanitation is a unique package and what we are facing at the moment is that we have difficulties, because, in order for your projects to make sense in the schools based on what we are teaching students, you cannot talk about water without talking about sanitation and vice versa” (Participant #1)

Frame analysis of the water problem in Maputo, Mozambique